Efficacy of psychostimulant drugs for cocaine dependence

Review question

We investigated whether psychostimulant substitution was safe and effective for treating patients with cocaine dependence.


Cocaine dependence is a frequent disorder for which no medication has been approved for treatment. Substitution therapy involves the replacement of the abused drug, which is often illegal and used several times a day, by a legal, orally administered and longer-acting one. A substitute drug has to have similar effects as the abused one, but with a lower addictive potential, enabling drug abstinence and patient adherence to medical and psychological assistance. This strategy can increase the abstinence rate in patients with heroin and tobacco dependence. In this review, we investigated whether psychostimulant substitution with medications that have psychostimulant effect was effective for treating patients with cocaine dependence.

Search date: the evidence is current to 15 February 2016.

Studies and participants' characteristics

We reviewed the evidence about the effect of psychostimulants on cocaine abstinence, safety and retention to treatment in patients with cocaine dependence. We found 26 studies that had enrolled 2366 participants and investigated the effects of psychostimulants against placebo for cocaine abuse or dependence. Most participants were men (75%) in their middle age (mean age 39.6 years). About half (47.6%) were African American, and 39.3% were white. The most common way they used cocaine was smoking. All but two studies took place in the USA, and they studied the effects of nine medications with a psychostimulant effect: bupropion, dexamphetamine, lisdexamfetamine, methylphenidate, modafinil, mazindol, methamphetamine, mixed amphetamine salts and selegiline. All clinical trials provided psychotherapy. Study length ranged from 6 to 24 weeks.

Key results

Investigators assessed cocaine abstinence (determined by urinalysis) in participants receiving the study intervention versus those receiving placebo. Though some analyses found that cocaine abstinence was higher with psychostimulants than with placebo, we are uncertain whether psychostimulants decrease cocaine use among participants who continue to use it or if they increase the number of people who stay clean, as the quality of the evidence was very low.

We also investigated the effect of the interventions studied on treatment retention. This outcome is important because withdrawing treatment and scheduled visits can suggest relapse to cocaine use. Psychostimulants probably make little or no difference when compared with placebo (moderate quality of evidence)

Psychostimulants appear well tolerated and are not associated with serious adverse events. Furthermore, psychostimulants show more favourable outcomes for some groups of patients, such as methadone-maintained, dual heroin-cocaine addicts, for whom there were positive results on both cocaine and heroin use.

Quality of the evidence

We did not consider any study to be free from risk of bias. We judged the quality of evidence to be very low for the outcomes of cocaine use and sustained abstinence but moderate for retention in treatment.

University researchers performed all studies with public funding, although eight of them also had additional private funding.


The efficacy of psychostimulants for cocaine dependence is not entirely clear, but these treatments appear promising and deserve further investigation.

Authors' conclusions: 

This review found mixed results. Psychostimulants improved cocaine abstinence compared to placebo in some analyses but did not improve treatment retention. Since treatment dropout was high, we cannot rule out the possibility that these results were influenced by attrition bias. Existing evidence does not clearly demonstrate the efficacy of any pharmacological treatment for cocaine dependence, but substitution treatment with psychostimulants appears promising and deserves further investigation.

Read the full abstract...

Cocaine dependence is a severe disorder for which no medication has been approved. Like opioids for heroin dependence, replacement therapy with psychostimulants could be an effective therapy for treatment.


To assess the effects of psychostimulants for cocaine abuse and dependence. Specific outcomes include sustained cocaine abstinence and retention in treatment. We also studied the influence of type of drug and comorbid disorders on psychostimulant efficacy.

Search strategy: 

This is an update of the review previously published in 2010. For this updated review, we searched the Cochrane Drugs and Alcohol Group Trials Register, CENTRAL, MEDLINE, Embase and PsycINFO up to 15 February 2016. We handsearched references of obtained articles and consulted experts in the field.

Selection criteria: 

We included randomised parallel group controlled clinical trials comparing the efficacy of a psychostimulant drug versus placebo.

Data collection and analysis: 

We used standard methodological procedures expected by Cochrane.

Main results: 

We included 26 studies involving 2366 participants. The included studies assessed nine drugs: bupropion, dexamphetamine, lisdexamfetamine, methylphenidate, modafinil, mazindol, methamphetamine, mixed amphetamine salts and selegiline. We did not consider any study to be at low risk of bias for all domains included in the Cochrane 'Risk of bias' tool. Attrition bias was the most frequently suspected potential source of bias of the included studies. We found very low quality evidence that psychostimulants improved sustained cocaine abstinence (risk ratio (RR) 1.36, 95% confidence interval (CI) 1.05 to 1.77, P = 0.02), but they did not reduce cocaine use (standardised mean difference (SMD) 0.16, 95% CI −0.02 to 0.33) among participants who continued to use it. Furthermore, we found moderate quality evidence that psychostimulants did not improve retention in treatment (RR 1.00, 95% CI 0.93 to 1.06). The proportion of adverse event-induced dropouts and cardiovascular adverse event-induced dropouts was similar for psychostimulants and placebo (RD 0.00, 95% CI −0.01 to 0.01; RD 0.00, 95% CI −0.02 to 0.01, respectively). When we included the type of drug as a moderating variable, the proportion of patients achieving sustained cocaine abstinence was higher with bupropion and dexamphetamine than with placebo. Psychostimulants also appeared to increase the proportion of patients achieving sustained cocaine and heroin abstinence amongst methadone-maintained, dual heroin-cocaine addicts. Retention to treatment was low, though, so our results may be compromised by attrition bias. We found no evidence of publication bias.

Health topics: