An Open Trial of Sustained Release Bupropion for Attention-Deficit/Hyperactivity Disorder in Adults with ADHD plus Substance Use Disorders

Abstract

Objective

To evaluate the effectiveness and tolerability of bupropion in adults with Attention-Deficit/Hyperactivity Disorder (ADHD) and comorbid active Substance Use Disorders (SUD).

Methods

This was a six-week open trial of sustained-release (SR) bupropion in adults aged 18 to 55 years diagnosed with both ADHD and SUD. Bupropion-SR was initiated at 100 mg SR and increased weekly to a target dose of 200 mg SR twice daily. Subjects were assessed on multiple outcomes including ADHD, SUD, and adverse effects. All analyses were intent to treat, with last observation carried forward.

Results

Thirty-two subjects were treated with bupropion, with nineteen subjects completing the entire protocol (59%). At end point there were clinically significant reductions in the ADHD RS (34.1±8.2 to 19.4±11.4, −43%, t=6.49, p<0.0001) and the Clinical Global Impression (CGI) of ADHD severity (baseline=5.0, endpoint=3.8, −24%, t=6.16, p<0.0001). In contrast, there were clinically negligible effects on the self-report of substance use (p’s >0.05) and on the overall CGI of SUD severity (−23%, t=4.95, p<0.0001).

Conclusions

Results from this open trial suggest that in adults with ADHD and SUD, treatment with bupropion-SR is associated with clinically significant reductions in ADHD, but not SUD.

Introduction

Attention-Deficit/Hyperactivity Disorder (ADHD) is increasingly recognized to persist from childhood, through adolescence and into adulthood.^{1, 2} Adults with ADHD have been shown to be at increased risk for a variety of comorbid disorders^3–7 including substance use disorders (SUD, including alcohol and/or drug abuse or dependence). Overall, approximately 50% of untreated adults with ADHD have a lifetime history of SUD.^{8, 9} Within ADHD the most frequently reported non-nicotine substances of abuse include marijuana (illicit drug) and alcohol.⁸

In addition to increasing the risk for developing SUD, persistence of ADHD has been shown to lead to an earlier onset of SUD in adolescents, more severe SUD, a longer duration of SUD, and lower rates of remission from SUD^8–13. Conversely, high rates of ADHD have also been reported in adults with SUD.^{6, 14, 15} Studies of both epidemiologically derived and clinically referred adults with SUD find that between 10–25% of these adults have ADHD^{6, 14–16} and that ADHD predicts a poorer response to treatment for SUD.^{14, 15}

Adults with ADHD and SUD present special challenges to clinicians. While consensus exists regarding the need for intervention in adults with ADHD and SUD, important questions persist in the management of these patients.^17–19 For instance, should ADHD be treated simultaneously with the SUD? Will ADHD pharmacotherapy result in reduced SUD? Another complicating factor is that clinicians are reluctant to utilize stimulant medications in substance abusing groups. Unfortunately, a limited literature exists that describes the assistance in the clinical management of these cases who are systematically excluded from clinical trials in adults with ADHD. Hence, studies of agents with low liability for abuse or diversion are necessary in this population.

One compelling medication for the treatment of adults with ADHD plus SUD is bupropion. Bupropion, an aminoketone antidepressant has indirect dopaminergic and noradrenergic enhancing effects.²⁰ Bupropion has been shown to be effective in the treatment of ADHD in adults.^21–23 Bupropion has also been shown to be effective in reducing impulse control disorders including nicotine cravings.²⁴

A limited literature exists on bupropion in the treatment of SUD in patients with comorbid ADHD. In a small open trial, Levin and colleagues²⁵ studied bupropion along with relapse prevention strategies in eleven adults with ADHD and comorbid cocaine abuse. While retention in the study was high (80%), significant improvements in ADHD were observed, and cocaine cravings and use were significantly lower at the end of study, investigators could not definitively state that bupropion was significantly better than previous treatments. These improvements were of similar magnitude to the only controlled study of bupropion. In a well conducted, 12 week randomized controlled study of bupropion and MPH in 98 methadone maintained subjects with ADHD receiving cognitive behavioral therapy, Levin et al.²⁶ reported nonsignificant improvements in ADHD in those receiving bupropion and MPH compared to placebo and methylphenidate. Similarly, nonsignificant improvement in cocaine use was reported across all groups.²⁶ While very useful, the generalization of these findings to adults with less severe SUD (e.g. marijuana) or those not engaged in active SUD treatment remains to be seen. We now report on the tolerability and efficacy of sustained-release (SR) bupropion in a sample of adults with ADHD and comorbid SUD. Based on the available literature, we hypothesized that bupropion-SR would be effective in reducing symptoms of ADHD and in reducing SUD.

Methods

Subjects

Subjects were outpatient men and women between 18 and 55 years of age who were recruited from advertisements and clinical referrals to a psychopharmacology outpatient clinic seeking treatment for ADHD. In addition to having ADHD these subjects were identified as having active non-nicotine SUD. We excluded potential subjects if they had any of the following: (1) clinically significant and unstable medical conditions, history of cardiac arrhythmias, eating disorders, seizure disorder, or hypertension; (2) clinically significant abnormal ECG; (3) pregnant or nursing females; (4) mental retardation (I.Q. <75); (5) organic brain disorders; (6) clinically unstable psychiatric conditions, including patients with active suicidal ideation, lifetime bipolar or psychotic disorders; (7) an unstable substance use disorder (i.e., demonstrating active withdrawal from substances meriting medical treatment, recent intravenous use, unable to utilize either public transportation to/from hospital or to ensure safety while driving); or (8) current use of psychotropic medication known to treat ADHD. Patients treated with naltrexone for decreasing alcohol cravings, methadone as opioid replacement, and nicotine patch or gum as tobacco replacement were admitted into the study. The institutional review board approved this study and all subjects completed a written informed consent prior to inclusion in the study.

Assessment Measures

Subjects underwent a standard clinical assessment comprising of a clinical psychiatric evaluation, a structured diagnostic interview, a cognitive battery, a medical history, and an electrocardiogram (ECG). The structured diagnostic interview used was the Structured Clinical Interview for DSM-IV (SCID),²⁷ supplemented for childhood disorders by unmodified modules from the Kiddie SADS E (Epidemiologic Version).²⁸

All cases were presented to a committee composed of board certified child psychiatrists and licensed psychologists. Diagnoses presented for review were considered positive only if the diagnosis would be considered clinically meaningful due to the nature of the symptoms, the associated impairment, and the coherence of the clinical picture. Discrepant reports were reconciled using the most severe diagnosis from any source unless the diagnosticians suspected that the source was not supplying reliable information. In addition to the assessment of abuse and dependence, subjects were queried for any use of illicit drugs during the SUD module of the structured interview. Early use of drugs or alcohol was not included in establishing diagnoses of conduct disorder.

To obtain a full diagnosis of adult ADHD, the subject must have (1) met full DSM-IV criteria for diagnosis of ADHD by the age of seven as well as currently (within the past month); (2) described a chronic course of ADHD symptomatology from childhood to adulthood and (3) endorsed a moderate or severe level of impairment attributed to the ADHD symptoms. Diagnostic reliability between raters and board certified child-psychiatrists was excellent. A Kappa of 1.0 was obtained for ADHD with a 95% confidence interval of 0.8–1.0.

The diagnosis of SUD was based on a clinical psychiatric interview using DSM-IV criteria. Adults diagnosed with current SUD had to demonstrate a maladaptive pattern of substance use resulting in one or more of the following: failure to fulfill major responsibilities, placing themselves in dangerous situations due to their substance use, experience substance-related legal problems and/or continue a pattern of recurrent substance use despite persistent social, interpersonal and/or legal problems. The diagnosis of SUD was confirmed by SCID for DSM-IV including an independent review by a board certified psychiatrist with additional credentials in SUD (TW) who was blind to the results of the clinical interview.²⁷

To assess the reliability of our diagnostic procedures, we computed kappa coefficients of agreement by having three experienced, board-certified child and adult psychiatrists diagnose subjects from audiotaped interviews made by the assessment staff. Based on 500 assessments from interviews of children and adults, the median kappa coefficient was 0.98. Kappa coefficients for individual diagnoses included: major depression (1.0), mania (0.95), attention deficit hyperactivity disorder (ADHD; 0.88), conduct disorder (CD; 1.0), oppositional defiant disorder (ODD; 0.90), antisocial personality disorder (ASPD; 0.80), and substance use disorder (1.0).

One of our primary outcomes was ADHD. To assess ADHD improvement, we used the ADHD Rating Scale (ADHD RS).^{29, 30} This psychometrically validated scale³¹ updated for DSM-IV,³⁰ assesses each of the 18 individual criteria symptoms using an identical severity grid (0=not present; 3=severe; overall minimum score=0; maximum score=54) which has been shown to be correlated with ADHD in adults³² and is medication sensitive.^{33, 34} An intra-class correlation of 0.85 was obtained for the ADHD RS.

The other primary outcome was self-reported substance use. At each visit, we used an eight item self-report scale derived from modules of the SCID and KSADS to measure substance use.^{27, 28} We used a modification of the time line follow-back method³⁵ in which subjects were administered a substance use questionnaire at each visit to cover the preceding week. The questionnaire asked whether subjects had used any of the following subjects in the prior week: alcohol, cocaine/crack, marijuana/pot, stimulants/uppers, LSD/mescaline, tranquilizers/benzos, pain killers, heroin/opiates, PCP, sniff gases or fumes, or other, and how often they used these substances.

For depression we used the Hamilton Depression Scale (HAM-D, minimum=0; max=84)³⁶ and the Beck Depression Inventory (BDI, minimum=0; max=63).³⁷ For anxiety, we used the Hamilton Anxiety Scale (HAM-A, minimum=0; max=56).³⁸

Similar to previous reports^{22, 34} overall severity for ADHD, substance use, depression and anxiety was assessed with the Clinical Global Impression (CGI) Scales.³⁹ The CGI includes Global Severity (CGI-S; 1=not ill, to 7= extremely ill) and Global Improvement (CGI-I; 1–2 very much to much improved compared to 3–6 mild improvement to very much worse). The intra-class correlation coefficient for the Clinical Global Impressions was 0.91.

While the ADHD RS and CGIs were administered at baseline and at each follow-up visit of the study, the HAM-A, HAM-D, and BDI were administered at baseline and at the end of study. Adverse experiences were by spontaneous report with systematic querying for their presence and were recorded at each visit.

Procedures

This was an open, prospective, six-week trial of sustained release bupropion (up to 200 mg BID) in adults with DSM-IV ADHD and concurrent SUD. Weekly supplies of bupropion were dispensed by the pharmacy in 100 mg tablets. Subjects were instructed to take their medication upon arising and again approximately six hours later. Compliance was monitored by pill counts at each physician visit. Study medication was begun at 100 mg SR in the morning and increased by 100 mg SR weekly in twice a day dosing up to 200 mg SR twice daily (week 4), unless adverse events emerged or the subject noted optimal improvement at a lower dose. Subjects were queried about any adverse events, with a specific focus on perceived interactions between bupropion and illicit substances. Blood pressure and pulse were measured at baseline and each subsequent visit. The study’s timeline was based on previous trials that demonstrated clinical efficacy at week six when subjects were titrated at four weeks ²².

Statistical Analysis

For the purposes of our main aim to examine the effectiveness and tolerability/safety of bupropion-SR, we included in our analyses all 32 subjects who began treatment. All analyses were intent to treat (at least one dose of bupropion) with last observation carried forward (LOCF). Differences between the ADHD RS, CGI severity scales and vital signs at baseline and endpoint were analyzed using paired t-tests. The non-parametric Wilcoxon Sign Rank test was used to measure the differences between HAM-A, HAM-D and BDI at baseline and endpoint. Pearson’s Chi Squared test was used to assess the influence of ADHD and SUD responses on the reciprocal outcome. We also used McNemar’s test to examine the differences in the substance questionnaire between baseline and endpoint. An alpha-level of 0.05 was used to assert statistical significance; all statistical tests were two-tailed. Data are presented as mean ± SD unless otherwise stated. We calculated all statistics using STATA 10.0.

Results

Subjects

Of the 122 subjects screened, 35 patients signed consent and enrolled in the study, and 32 subjects started study medications. Of those who did not begin treatments, 35 did not return for follow-up, 15 were not interested, 8 had current unstable severe substance abuse, 8 did not meet criteria for ADHD, 8 had bipolar or psychotic disorder, 6 had medical contraindications, 4 were receiving exclusionary psychotropics (treatment for ADHD), 3 had primary major depression, and 3 had previous exposure to bupropion.

Of the 35 patients who signed consent, 3 subjects dropped due to noncompliance, 32 subjects took study medication and 19 subjects (19/32; 59%) completed the six-week protocol. Among those who completed at least one week of treatment, 9 subjects dropped due to noncompliance or were lost to follow-up and 4 subjects withdrew due to adverse events including: elevated blood pressure, exacerbation of panic attacks, and a significant rash. The final sample of those receiving at least one dose of bupropion consisted of 26 men and 6 women who ranged in age from 20 to 53 years (mean age = 32 ±8.6 years). The majority of subjects were Caucasian. We examined the demographics, diagnoses, severity, and previous stimulant treatment between subjects who completed the study (N=19) and subjects who dropped (N=12). We found a significant difference for the current GAF with dropped subjects having a lower score (52.9 ± 2.4) compared to completed subjects (57.7 ± 4.2; t=−3.0, p=0.01). All other analyses were not significant (all p values > 0.05).

The type of SUD observed in the 26 completed SCIDs were largely mixed with 69% having combined alcohol plus drug abuse/dependence, 12% having alcohol abuse or dependence alone, and 12% having drug abuse or dependence alone. One subject did not meet SUD criteria based on the SCID, but was found to have a clinically significant level of substance abuse by clinician assessment. As depicted in Table 1, the most common lifetime comorbid psychiatric disorders in our sample included Conduct Disorder (54%), Major Depression (42%), Antisocial Personality Disorder (31%), and Oppositional Defiant Disorder (31%). Prior to entering this study, 8 subjects (31%) had received prior pharmacotherapy for ADHD. No subject was on naltrexone during the study. Despite average to above average intelligence (IQ 108±13.6) in this group of ADHD adults, 34% required tutoring in school and 19% repeated at least one grade.

Table 1.

Demographic and Clinical Characteristics of Sample (N=32)

	Mean ± SD
Age (Years)	32.0 ± 8.5
IQ*	108 ± 13.6
	%
Sex
Male	81
Previous ADHD Treatment**	31
Lifetime Comorbidity**
Attention-Deficit/Hyperactivity Disorder	100
Inattentive Subtype	35
Hyperactive/Impulsive Subtype	4
Combined Subtype	61
Substance Use Disorders***	100
Alcohol Dependence	34
Alcohol Abuse	63
Drug Dependence	28
Drug Abuse	66
Disruptive Disorders
Oppositional Defiant Disorder	31
Conduct Disorder	54
Mood Disorders
Major Depression	42
Anxiety Disorders
Panic Disorder	12
Agoraphobia	4
Social Phobia	23
Obsessive Compulsive Disorder	4
Generalized Anxiety Disorder	8
Separation Anxiety Disorder	0
Antisocial Personality Disorder	31
	Mean ± SD
Past GAF****	48 ± 4.0
Current GAF****	56 ± 4.3

^*N = 25 subjects with completed cognitive battery

^**N = 26 subjects with completed Structured Clinical Interview for DSM-IV

^***Diagnoses not mutually exclusive

^****Global Assessment of Functioning

Effect on ADHD

Using both dimensional and categorical definitions of ADHD improvement, bupropion was found to be clinically and statistically significant in reducing ADHD symptomatology. At baseline, subjects had a mean score of 34.1±8.2 on the ADHD RS. At endpoint, scores were significantly reduced by 43% (19.4±11.4, t=6.49, p<0.0001, Table 2). Of the entire sample 67% (21/32) had a 30% reduction in their ADHD RS. Similarly, ADHD CGI severity scores declined from the marked range at baseline (5.0±0.7) to a mild to moderate range at endpoint (3.8±1.1, t=6.16, p<0.0001, Table 2). Using predefined criteria of a CGI-Improvement of 1 or 2 (much to very much improved) for ADHD, 41% of subjects were considered improved at endpoint.

Table 2.

Effectiveness of Bupropion Sustained-Release in the Treatment of Adults with ADHD and Substance Use Disorders: Main Outcomes (N=32)

Outcome	Baseline (mean±S.D.)	End of Treatment* (mean±S.D.)	t	p
ADHD Rating Scale (ADHD RS)	34.1±8.2	19.4±11.4	6.49	<0.0001
Clinical Global Impression Scale of ADHD Severity (ADHD CGI-S)	5.0±0.7	3.8±1.1	6.16	<0.0001
Clinical Global Impression Scale of SUD Severity (SUD CGI-S)	4.0±0.9	3.1±0.8	4.95	<0.0001
			z	p
Hamilton Depression Scale (HAM-D)	8.1±6.5	4.5±6.9	2.97	0.003
Hamilton Anxiety Scale (HAM-A)	8.8±9.1	4.5±5.7	2.93	0.0034
Beck Depression Inventory (BDI)	8.9±5.9	3.7±5.1	3.50	0.0005

^*Intent to treat at least one dose of bupropion with last observation carried forward

Because of our relatively high attrition rate, we also analyzed our main outcomes based on completers (N=19) and found similar results. For the ADHD RS, scores were significantly reduced by 39% (t=8.26, p<0.0001) and for the CGI-S, scores declined from a marked range (5.1 ± 0.8) to a mild to moderate range at endpoint (3.3 ± 1.0, t=9.4, p<0.0001). For the CGI-I, 68% of completers were considered improved at endpoint.

Effect on SUD

No clinically significant reductions were observed in the group over the course of the study in either the self-report of SUD or the investigator rated CGI severity SUD scores (Table 2). At baseline (week 0), 97% (31/32) of subjects reported using substances in the previous week on the substance use questionnaire. Subsequent rates of self reported weekly use were 100% (32/32) at week 1, 100% (27/27) at week 2, 91% (20/22) at week 3, 90% (19/21) at week 4, 100% (18/18; one subject unable to attend due to unforeseen circumstances) at week 5, and 95% (18/19) at week 6. There was no clinically or statistically significant difference between the percent of subjects who reported substance use at baseline vs. the end of the study (χ²=1.00, p=0.32).

Using predefined criteria of a CGI-Improvement of 1 or 2 (much to very much improved) for SUD, 31% of subjects were considered improved at endpoint (out of 29 subjects completing CGI-I SUD). Using the CGI severity SUD scores, the group demonstrated negligible reductions in SUD during the course of bupropion treatment (baseline 4.0±0.9 and endpoint 3.1±0.8; 23% reduction; t=4.95; p<0.0001, Table 2).

For the completers, we found similar results. For the CGI-S, we found negligible reductions (baseline 3.9 ± 1.0 and endpoint 2.8 ± 0.8; 29% reduction; t=4.60; p<0.001) and for the CGI-I, we found that 42% of completers were considered improved at endpoint.

Impact of ADHD improvement on SUD outcome

We further evaluated the influence of ADHD and SUD responses on the reciprocal outcome. We first examined ADHD responders (CGI-I for ADHD of 1 or 2 [much or very much], N=12) and found they were more likely to have improved SUD at endpoint measured by the CGI-S for SUD (≤ 3 [mildly ill] vs. >3, N=11 (92%) vs. N=1 (8%), χ²(1) = 6.08, p=0.019). Conversely, we examined the few SUD responders (CGI-I for SUD of 1 or 2, N=9) and found they were more likely to have improved ADHD at endpoint (CGI-S for ADHD≤ 3 [mildly ill] vs. >3; N=7 (78%) vs. N=2 (22%), χ²(1) = 7.13, p=0.014).

Ratings of Anxiety and Depression

Although our group had high rates of lifetime Anxiety Disorders (see Table 1), baseline (8.1±6.5) and endpoint (4.5±6.9) ratings of current anxiety, as measured by the HAM-A (N=19), were relatively low (44% reduction over 6 weeks; z=2.93; p=0.003, Table 2). Similarly, although 42% of our sample had lifetime histories of Major Depression, baseline (8.8±9.1) and endpoint (4.5±5.7) ratings of current depression, as measured by the HAM-D (N=19), were minimal (49% reduction; z=2.93; p=0.0034, Table 2). Of the 20 subjects that completed the BDI, there were mild reductions in depression during the course of the bupropion treatment (baseline 8.9±5.9 and endpoint 3.7±5.1; 59% reduction; z=3.50; p=0.0005, Table 2). The reductions in both anxiety and depression were not felt to be clinically significant.

Adverse Effects

There were no serious adverse drug reactions during the trial. Adverse effects observed were mild to moderate in nature, the most common (>5% of subjects) included appetite loss, insomnia, gastro-intestinal problems, headaches and irritability. Interestingly, no adverse effects related to interactions with substances of abuse were reported. There were no clinically meaningful changes in cardiovascular signs with the treatment.

Discussion

The results from this prospective open evaluation of bupropion in ADHD plus SUD adults show an improvement in ADHD with only meager and clinically insignificant, while statistically significant, improvement in investigator rated, but not self-reported SUD. A trend emerged in that adults who responded favorably to bupropion for ADHD reduction had improvement in their SUD; conversely, adults with improved SUD at outcome manifested improvement in their ADHD. While bupropion was relatively well tolerated, 41% of the sample dropped out of the study prematurely largely due to a failure to return for study visits. The limited improvement in ADHD coupled with the poor completion rate furthers the notion of stabilizing SUD if possible prior to treating ADHD pharmacologically.

Our findings are similar to those of Riggs et al.⁴⁰ in adolescents treated with bupropion in which reductions in ADHD were noted; however, no change in SUD was observed. Similarly, like Levin et al.²⁵ who studied a small sample of adult outpatients with ADHD plus cocaine addiction openly, we observed significant reductions in ADHD symptoms. Unlike Levin et al.²⁵ who observed improvement in cocaine craving and use, we only found a negligible reduction in SUD. Of interest, in a subsequent controlled trial in adults with cocaine abuse who were receiving methadone maintenance, bupropion was not associated with improvement in either ADHD or SUD.²⁶ Of interest, in the latter Levin study,²⁶ the high placebo response for ADHD may have negated the significance of the ADHD signal.

The current findings also mirror a larger and growing literature with other ADHD agents in comorbid SUD. For example, placebo controlled trials with stimulants in comorbid ADHD plus SUD demonstrate variable improvement in ADHD without significant evidence of improved SUD.^{26, 41–46} For example, Shubiner et al.⁴¹ reported on results from 48 ADHD adults with cocaine addiction receiving placebo or methylphenidate for 12 weeks who were concomitantly receiving group and individual cognitive behavioral therapy. While clinically significant reductions in ADHD were found, there were no differences in self-report cocaine use, urine toxicology or cocaine craving. Similarly, data by Riggs et al.⁴² from a placebo controlled trial of pemoline for adolescents with ADHD, failed to demonstrate improved outcome for ADHD or SUD. Carpentier et al.⁴³ in adults receiving methylphenidate, and Levin et al. in two studies of methylphenidate^{26, 44} in adults found some variable reduction in ADHD symptoms without notable improvements in SUD. In contrast, Wilens et al. found that recently abstinent alcoholics with ADHD showed clinically and statistically significant improvements in both ADHD and in recurrent heavy drinking, but not relapse, in those randomized to atomoxetine compared to placebo.⁴⁷ Hence, the controlled trials along with larger open studies seem to suggest that treatment of individuals with ADHD plus current active SUD while benefiting ADHD has only minimal effects on SUD reduction and are plagued by issues of retention. In contrast, ADHD pharmacotherapy in very recently abstinent SUD may be associated with improved ADHD and SUD.⁴⁷

The high attrition rate in the current study is noteworthy. Forty-one percent of the current sample dropped out of the study prematurely. Similar findings of relatively high attrition in studies of subjects with ADHD plus SUD in virtually all studies have been reported.⁴⁵ Given that retention is associated with successful SUD and overall outcome, the extant literature supports the notion that if possible, the addiction needs to be addressed prior to initiation of ADHD pharmacotherapy to maximize improvement in all domains.^{47, 48}

The notion of ADHD response as a function of psychiatric comorbidity remains an issue of active debate. Studies, for instance, with mood disorders seem to demonstrate improved ADHD response in those individuals in whom the mood disorder comorbidity is adequately treated.⁴⁹ While the overall effect of our current data indicates ADHD response in comorbid SUD, we found that adults with improvement in their ADHD or SUD were more likely to manifest improvement in their reciprocal disorder compared to those adults with continued ADHD or SUD at endpoint. Therefore, while the aggregate literature examining treatment with either stimulants or antidepressants demonstrates that ADHD symptoms improve in the context of comorbid SUD, further studies assessing the effect of each disorder on the outcome of the other disorder are necessary.

Bupropion was relatively well tolerated in the current study with no serious adverse effects and no notable drug interactions with substances of abuse. These findings parallel other studies with bupropion^{26, 50} signaling good tolerability in context to SUD and other disorders. Our findings stand in contrast to observations of Haney et al.⁵¹ who observed significant worsening of mood during marijuana withdrawal in subjects concomitantly treated with bupropion. Despite our inquiry into clinically significant adverse events with bupropion taken in juxtaposition to substances of abuse, our small sample size precludes fully assessing safety in this group and clinicians must continue to use caution when using this medication in the context of SUD.

The results of this study must be viewed in the context of their substantial methodological limitations. This was an open study and thus susceptible to observer bias. Improvements may have been confounded by participation in the study and not actual medication effect (i.e., placebo effect) as well as regression to the mean. While our procedural timeline was based on previous studies of substance abuse and ADHD, the study duration was only six weeks long with subjects receiving the full dose at week four. This time period may not have been adequate to show substance use improvements or allow for sufficient exposure to medication. The generalizability of our sample is compromised due to the high attrition rate, the short duration of the study, and the limited racial and socioeconomic heterogeneity of the sample. Our reports of substance use were by report from our subjects. Given that essentially all subjects reported substance use during the study, we do not believe that our study would have benefited from objective testing; and self report is a viable method of SUD outcome.³⁵ The HAM-A and HAM-D were low at baseline and with treatment and as such, we were unable to accurately examine relationships with ADHD. Because of the low scores for both anxiety and depression, we do not believe these comorbid disorders affected the overall results. Lastly, we did not obtain neuropsychological data.

Despite these limitations, the results of this open study in adults with ADHD plus SUD suggest that bupropion improved ADHD, but only had a meager effect on SUD. The results coupled with the high attrition rate gives credence to a growing notion of the need to sequence the SUD intervention, at least briefly, prior to pharmacotherapy for ADHD in individuals with the combined comorbidity. Further controlled studies of bupropion and other anti-ADHD agents in newly abstinent individuals with ADHD plus SUD are warranted as well as treatment sequencing studies in those with an active SUD.