Psicoestimulantes para Depreion

Psychostimulants for depression (Review)
Candy B, Jones L, Williams R, Tookman A, King M
This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library 2009, Issue 1 http://www.thecochranelibrary.com
Psychostimulants for depression (Review) Copyright 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
TABLE OF CONTENTS HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . AUTHORS CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.1. Comparison 1 Psychostimulants versus placebo, Outcome 1 Reduction in depression symptoms (shortterm). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.2. Comparison 1 Psychostimulants versus placebo, Outcome 2 Reduction in depression symptoms (medium term). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.3. Comparison 1 Psychostimulants versus placebo, Outcome 3 Clinical response (short term) - any measure/denition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.4. Comparison 1 Psychostimulants versus placebo, Outcome 4 Improvement in cognitive functioning (short term). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.5. Comparison 1 Psychostimulants versus placebo, Outcome 5 Reduction in fatigue (short term). . . . Analysis 1.6. Comparison 1 Psychostimulants versus placebo, Outcome 6 Reduction in fatigue (medium term). . . Analysis 1.7. Comparison 1 Psychostimulants versus placebo, Outcome 7 Reduction in hypersomnia (medium term): Epworth Sleepiness Scale. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.8. Comparison 1 Psychostimulants versus placebo, Outcome 8 Improvement in health-related quality of life (short term). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.9. Comparison 1 Psychostimulants versus placebo, Outcome 9 Acceptability: dropouts due to side effects (short term). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.10. Comparison 1 Psychostimulants versus placebo, Outcome 10 Tolerability: proportion of participants experiencing side effects (short term). . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.11. Comparison 1 Psychostimulants versus placebo, Outcome 11 Tolerability: proportion of participants experiencing side effects (medium term). . . . . . . . . . . . . . . . . . . . . . . . . Analysis 1.12. Comparison 1 Psychostimulants versus placebo, Outcome 12 Tolerability: dizziness (short term). . . Analysis 1.13. Comparison 1 Psychostimulants versus placebo, Outcome 13 Tolerability: drowsiness (short term). . Analysis 1.14. Comparison 1 Psychostimulants versus placebo, Outcome 14 Tolerability: headache (short term). . . Analysis 1.15. Comparison 1 Psychostimulants versus placebo, Outcome 15 Tolerability: insomnia (short term). . . Analysis 1.16. Comparison 1 Psychostimulants versus placebo, Outcome 16 Tolerability: nausea (short term). . . . Analysis 1.17. Comparison 1 Psychostimulants versus placebo, Outcome 17 Tolerability: tremor (short term). . . . Analysis 2.1. Comparison 2 Psychostimulants versus antidepressants, Outcome 1 Reduction in depression symptoms (short term). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 2.2. Comparison 2 Psychostimulants versus antidepressants, Outcome 2 Tolerability: number of participants experiencing side effects (medium term). . . . . . . . . . . . . . . . . . . . . . . . . Analysis 3.1. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 1 Reduction in depression symptoms (short term). . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 3.2. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 2 Reduction in depression symptoms (medium term). . . . . . . . . . . . . . . . . . . . . . . . . Analysis 3.3. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 3 Clinical response (short term). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 3.4. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 4 Reduction in fatigue symptoms (short term). . . . . . . . . . . . . . . . . . . . . . . . . . . .
1 1 2 2 3 3 7 12 12 12 13 18 35 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61
i
Analysis 3.5. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 5 Reduction in fatigue symptoms (medium term). . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 3.6. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 6 Reduction in hypersomnia (short term): Epworth Sleepiness Scale. . . . . . . . . . . . . . . . . . . . Analysis 3.7. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 7 Reduction in hypersomnia (medium term): Epworth Sleepiness Scale. . . . . . . . . . . . . . . . . . . Analysis 3.8. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 8 Acceptability: dropouts due to side effects (short term). . . . . . . . . . . . . . . . . . . . Analysis 3.9. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 9 Acceptability: dropouts due to side effects (medium term). . . . . . . . . . . . . . . . . . . Analysis 3.10. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 10 Acceptability: dropouts due to lack of efcacy (medium term). . . . . . . . . . . . . . . . . . Analysis 3.11. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 11 Tolerability: proportion of participants experiencing side effects (short term). . . . . . . . . . . . . Analysis 3.12. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 12 Tolerability: headache (medium term). . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 3.13. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 13 Tolerability: insomnia (short term). . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 3.14. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 14 Tolerability: insomnia (medium term). . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 3.15. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 15 Tolerability: nausea (short term). . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 3.16. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 16 Tolerability: nausea (medium term). . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 3.17. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 17 8Tolerability: tremor (short term). . . . . . . . . . . . . . . . . . . . . . . . . . . . Analysis 3.18. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 18 Tolerability: upper respiratory tract infection (medium term). . . . . . . . . . . . . . . . . . Analysis 4.1. Comparison 4 Psychostimulants versus placebo: concomitant serious medical condition, Outcome 1 Reduction in depression symptoms (short term). . . . . . . . . . . . . . . . . . . . . . . ADDITIONAL TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . WHATS NEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . HISTORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . INDEX TERMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 76 83 83 83 83 84 84
ii
[Intervention Review]
Psychostimulants for depression

Bridget Candy1 , Louise Jones2 , Rachael Williams2 , Adrian Tookman2 , Michael King2
1 Marie Curie Palliative Care Research Unit, Department of Mental Health Sciences, Royal Free & University College Medical School , London, UK. 2 Marie Curie Palliative Care Research Unit, Department of Mental Health Sciences, Royal Free & University College Medical School, London, UK
Contact address: Bridget Candy, Marie Curie Palliative Care Research Unit, Department of Mental Health Sciences, Royal Free & University College Medical School , Hampstead Campus , Rowland Hill Street , London, NW3 2PF , UK. b.candy@medsch.ucl.ac.uk. bridget@metaclarity.com. Editorial group: Cochrane Depression, Anxiety and Neurosis Group. Publication status and date: Edited (no change to conclusions), published in Issue 1, 2009. Review content assessed as up-to-date: 10 February 2008. Citation: Candy B, Jones L, Williams R, Tookman A, King M. Psychostimulants for depression. Cochrane Database of Systematic Reviews 2008, Issue 2. Art. No.: CD006722. DOI: 10.1002/14651858.CD006722.pub2. Copyright 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
ABSTRACT Background Depression is common, disabling, costly and under-treated. There are problems in the current rst-line drug treatment, antidepressants, for moderate or severe depression. There is a body of research that has evaluated the effect of psychostimulants (PS) in the treatment of depression. This has not been reviewed systematically. Objectives To determine the effectiveness of PS in the treatment of depression and to assess adverse events associated with PS. Search methods Databases CCDANCTR-Studies and CCDANCTR-References were searched on 21/6/2006. Cochrane Central Register of Controlled Trials, MEDLINE, EMBASE, PsycInfo, AMED, CINAHL, Dissertation Abstracts and the National Health Service Research Register were searched. Selection criteria Randomised controlled trials (RCTs) assessing the effectiveness of PS were included. The trial population comprised adults of either sex with a diagnosis of depression. Data collection and analysis Two review authors extracted the data independently and assessed trial quality. Meta-analysis was considered for trials with comparable key characteristics. The primary outcome was depression symptoms, based on a continuous outcome, using the standardised mean difference (SMD), or a dichotomous measure of clinical response, using odds ratios (OR), with 95% condence intervals (CI). Main results Twenty-four RCTs were identied. The overall quality of the trials was low. Five drugs were evaluated; dexamphetamine, methylphenidate, methylamphetamine, pemoline and modanil. Modanil was evaluated separately as its pharmacology is different to that of the other PS. PS were administered as a monotherapy, adjunct therapy, in oral or intravenous preparation and in comparison with a placebo or an active therapy. Most effects were measured in the short term (up to four weeks). Thirteen trials had some usable
Psychostimulants for depression (Review) Copyright 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. 1
data for meta-analyses. Three trials (62 participants) demonstrated that oral PS, as a monotherapy, signicantly reduced short term depressive symptoms in comparison with placebo (SMD -0.87, 95% CI -1.40, -0.33, with non-signicant heterogeneity. A similar effect was found for fatigue. In the short term PS were acceptable and well tolerated. Tolerance and dependence were under evaluated. No statistically signicant difference in depression symptoms was found between modanil and placebo. Authors conclusions There is some evidence that in the short-term, PS reduce symptoms of depression. Whilst this reduction is statistically signicant, the clinical signicance is less clear. Larger high quality trials with longer follow-up and evaluation of tolerance and dependence are needed to test the robustness of these ndings and, furthermore, to explore which PS may be more benecial and in which clinical situations they are optimal.
PLAIN LANGUAGE SUMMARY Psychostimulants for depression Depression is common and under-treated. The current rst-line drug treatment for moderate or severe depression is antidepressants, but there are problems with their use. This review evaluated the evidence from randomised controlled trials (RCTs) evaluating the effectiveness of psychostimulants (PS) in the treatment of depression. Twenty-four RCTs were identied, of which 14 had data for meta-analysis. Five drugs were evaluated, including dexamphetamine, methylphenidate, methylamphetamine, pemoline and modanil. Modanil was evaluated separately as its pharmacology differs from other PS. Three small trials of PS involving a total of 62 participants indicated that oral treatment with PS in the short term (up to four weeks) signicantly reduced depressive symptoms when compared with placebo, however, the overall quality of the trials was poor, limiting condence in the ndings. Two trials involving 411 participants compared modanil against placebo when combined with antidepressant treatment at 6-8 weeks, and showed a non-signicant difference in reducing depression symptoms. One small trial of 50 participants compared oral modanil against placebo after 12 weeks of treatment, and also showed a non-signicant difference in reducing depression symptoms. No trials examined the longer-term effect of PS. Further well conducted trials with long term follow-up are required to nd out which PS may be more effective in the treatment of depression, and whether PS are more effective in certain subgroups of depressed patients.
BACKGROUND
Description of condition Depression is a common (Singleton 2001), disabling (Spitzer 1995; Cassano 2002), costly (Murray 1997; Thomas 2003) and under-treated condition (Bebbington 2000; Demyttenaere 2004; Singleton 2001; Wang 2005). The health burden of depression is recognised by the World Health Organisation (WHO) where an international initiative that aims to raise awareness, and improve recognition and treatment of depression has been launched. At a national level, several countries, including European states, the US, and Australia, have programme objectives in depression set as a priority for health in the 21st century. Depression is characterised by depressed mood and/or loss of interest or pleasure in nearly all activities in the presence of other symptoms such as loss of appetite, fatigue and lack of energy, sleep disturbance, restlessness or irritability, feelings of worthlessness or inappropriate guilt, difculty thinking, concentrating or making decisions and thoughts of death or suicide or attempts at suicide (APA 2005). Only about 30% to 40% of people in the community who meet diagnostic criteria for depression receive any form of medical or psychological treatment (Lin 1999; Singleton 2001) and this falls to 25% or less when only treatments judged to be of minimal adequacy or better are included (Young 2001). Talking therapies are popular and effective but are in short supply in community settings in Europe, USA and elsewhere. Description of intervention Depression is associated with disturbances in the neurotransmitter systems/serotonin, noradrenaline and/or dopamine in the central nervous system. Antidepressant drugs elevate the levels of these neurotransmitters in the synaptic cleft by blocking their re-uptake; all have been shown to be effective in the alleviation of depressive symptoms. However, there is still considerable doubt about the precise role of the monoamine neurotransmitters in depression (Delgado 2000). There has also been a reluctance to use other chemicals, such as PS, to increase monoamine concentrations and
2
function, usually because of fears of misuse or dependence. The current rst line drug treatment for moderate or severe depression is antidepressants such as selective serotonin reuptake inhibitors (e.g. uvoxamine, sertraline, citalopram) or tricyclic and related antidepressants (e.g. amitriptyline, imipramine, lofepramil) (NICE 2004). Monoamine oxidase inhibitors (e.g. phenelzine, moclobemide) may also be useful. However, there are a number of problems with their use: (i) There is evidence for the efcacy of antidepressant drugs but it has been argued that their clinical effect may be weak (Moncrieff 2005); (ii) adherence to treatment may be poor (Maddox 1994) and approximately one third of patients who adhere to treatment do not recover; (iii) antidepressants usually have a therapeutic effect within a week and possibly up to three weeks to have full effect; (iv) reduced energy is a core feature of depression but many antidepressants do not alleviate this symptom; (v) fatigue may also be a complication of therapy with antidepressants: this may be particularly important in people with co-morbid physical illness. How the intervention might work Psychostimulants (PS) are drugs known to reduce fatigue and promote alertness and wakefulness. They are a possible alternative to antidepressants, although the conditions for which they are used are limited. Currently in the UK they are only approved for use in the treatment of attention-decit hyperactive disorder (BNF 2007). With the exception of modanil and caffeine, PS increase synaptic activity of the monoamine neurotransmitters, dopamine, noradrenaline and serotonin and have an indirect agonist effect by increasing neurotransmitter function/release rather than a direct effect on postsynaptic receptors. All produce a similar increase in monoamine activity. The main advantage of PS over selective serotonin reuptake inhibitors and tricyclic antidepressants may be their time course of action. PS appear to act quickly: some patients report an improved sense of well being within 24 hours. They may also promptly reduce fatigue and increase energy, benets that may be crucial in those with severe physical illness or in older people (Nguyen 2005). Why it is important to do this review Why PS are not considered a treatment option in depression (NICE 2004) may relate in part to uncertainty about their effectiveness and safety. Earlier reviews of the evidence on effectiveness of PS for depression have reported inconsistent results across studies (Chiarello 1987; Linet 1989; Masand 1998;Orr 2007; Postolache 1999; Satel 1989). In addition there is potential for misuse, dependence and associated withdrawal reactions and under current legislation in the United Kingdom most PS are controlled substances. We believe it is time to review the place of PS in the treatment of depression and to decide whether further research is needed to evaluate their efcacy in particular settings, such as treatment for mood disorders and fatigue in patients with advanced progressive disease.
OBJECTIVES
1. To examine the effectiveness of PS in the treatment of depression. 2. To assess adverse events associated with use of PS as a treatment for depression.
METHODS
Criteria for considering studies for this review
Types of studies Randomised controlled trials (RCTs) were included. Observational cohort or cross-sectional studies were not included. For any RCTs using a cross-over design, we used data from analysis from the rst comparative phase only or where the means and standard deviations were reported for each trial arm separately. Types of participants Patient characteristics and diagnosis Participants were adults (aged 16 and over) of either sex receiving a psychostimulant (PS) as a treatment for depression. Trials were included if the diagnosis of depression was made according to any edition of the Diagnostic and Statistical Manual for Mental Health (DSM) or the World Health Organisations International Classication of Diseases (ICD) or when a clinician made the diagnosis. We undertook such a broad denition of depression, as we were aware that some trials evaluating the effect of PS predate the regular use of standardised diagnostic criteria for depression. Exclusion criteria Trials whose primary focus was to evaluate treatments in patients with substance dependence were excluded. Trials with children were not included. Comorbidity Patients with serious concomitant medical illness were included in the review. It was planned that trials with patients with co-morbid psychiatric diagnoses would also be included. Types of interventions Intervention Trials were included if they evaluated the effectiveness of any type of psychostimulant (PS) for any type of depression. Trials using the PS modanil were evaluated separately as its pharmacology is completely different from that of the other PS. Control conditions PS were evaluated against any type of control or comparison treatment or wait list group.
3
Main comparisons Outcomes were presented by comparison as follows: 1. PS as monotherapy versus placebo 2. PS as monotherapy versus other treatment (medication, psychological therapy) 3. PS versus other treatment as adjunctive treatment
Increased anxiety or depression Suicidal thoughts Acts of self-harm Outcomes were presented according to length of treatment, as follows: 1. short term (4 weeks or less) 2. medium term (5 to 12 weeks) 3. long term (13 weeks or longer).
Types of outcome measures Primary outcome The primary outcome measure was the effect of the PS on depressive symptoms or diagnosis using: 1. continuous measures, such as Hamilton Depression Scale (HAM-D) (Hamilton 1960) or Montgomery-Asberg Scale ( Montgomery 1979) 2. dichotomous measures - the proportion of people who respond to treatment: including a categorisation of the HAM-D score (Hamilton 1960) or another depression scale such as the Montgomery-Asberg Scale (Montgomery 1979) or any other validated depression scale into a 50% response or less. Secondary outcomes 1. Changes in other symptoms associated with depression including: changes in weight when not dieting or weight gain or changes in appetite; sleep disturbance (insomnia or hypersomnia (daytime sleepiness)); psychomotor agitation or retardation; fatigue or loss of energy; diminished ability to think or concentrate or recurrent thoughts of death or suicidal ideation (DSM-IV features of Depressive Disorders). Such symptoms were captured in validated scales such as for example the Montgomery-Asberg Scale (Montgomery 1979). 2. Remission criteria. 3. Social adjustment and functioning, such as the Social Adaptation Self-Evaluation Scale (Bosc 1997) 4. Health related quality of life, measured using the SF-36 (Ware 1992) 5. Acceptability - this was evaluated by: a). The total number of participants who dropped out due to inefcacy b). The total number of participants who dropped out due to side effects 6. Tolerability - this was evaluated by: a) Proportion of randomised participants experiencing some side effects b) Proportion of randomised participants experiencing the following side effects; Appetite changes Insomnia Reduced energy PS misuse and dependence Withdrawal reactions Tolerance Over arousal
Search methods for identication of studies

See: Depression, Anxiety and Neurosis Group methods used in reviews. 1. Electronic databases: The CCDAN registers was searched as below: a) CCDANCTR-Studies (searched on 21/6/2006) Diagnosis = Depress* or Dysthymi* or Adjustment Disorder* or Mood Disorder* or Affective Disorder* or Affective Symptoms and Intervention = amylobarbitone or Caffeine or dextroamphetamine or *amphetamine or methylphenidate or phenmetrazine or amoxetine or dexamfetamine or cocaine or phenylpropanolamine or pemoline or ephedrine or modanil or methylene or Psychostimulant*) CCDANCTR-References (searched on 21/6/2006) Keyword = Depress* or Dysthymi* or Adjustment Disorder* or Mood Disorder* or Affective Disorder* or Affective Symptoms and Free-Text = amylobarbiton* or caffeine or *amphetamine or methylphenidat* or phenmetrazin* or amoxetin* or dexamfetamin* or cocaine or phenylpropanolamin* or pemolin* or ephedrin* or modanil or methylen* or psychostimulant* b) The following databases were searched: MEDLINE PSYINFO EMBASE AMED CINHAL The Cochrane Central Register of Controlled Trials (CENTRAL) Dissertation Abstracts National Health Service Research Register. Search strategy used to search MEDLINE a) PS 1) psychostimulants or stimulants 2) central nervous system stimulants (EXPLODE MeSH), all subheadings) 3) dextroamphetamine (EXPLODE MeSH, all subheadings) 4) amphetamine (EXPLODE MeSH, all subheadings) 5) methylphenidate (EXPLODE MeSH, all subheadings)
4
6) phenmetrazine (EXPLODE MeSH, all subheadings) 7) dexamphetamine 8) dexedrine or methylphenidate hydrochloride 9) methylphenidate hydrochloride (EXPLODE MeSH, all subheadings) 10) cocaine (EXPLODE MeSH, all subheadings) 11) phenylpropanolamine (EXPLODE MeSH, all subheadings) 12) ritalin or cephalon or concerta or modanil or provigil or cocaine or seleguine 13) amphetamine* or diethylpropion or phenmetrazine or phendimetrazine or phenylpropanolamine 14) ephedrine (EXPLODE MeSH, all subheadings) 15) modanil 16) Caffeine 17) Central Nervous System Stimulants (EXPLODE MeSH, all subheadings) 18) #1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12 or #13 or #14 or #15 or #16 or #17. b) DEPRESSION 19) depressi* 20) depressed near patient* 21) depressed near symptom* 22) depressi* adj3 disorder* 23) depressi* adj3 symptom* 24) affective disorder* 25) mood adj3 disorder* 26) #19 or #20 or #21 or #22 or #23 or #24 or #25 c) FILTER FOR RCT 27) random* in ti, ab 28) factorial* in ti, ab 29) (crossover* or cross over* or cross-over*) in ti, ab 30) placebo* in ti, ab 31) (doubl* near blind*) in ti, ab 32) (singl* near blind*) in ti, ab 33) assign* in ti, ab 34) allocat* in ti, ab 35) volunteer* in ti, ab 36) crossover procedure (EXPLODE MeSH, all subheadings) 37) double blind procedure (EXPLODE MeSH, all subheadings) 38) randomized controlled trial (EXPLODE MeSH, all subheadings) 39) single blind procedure (EXPLODE MeSH, all subheadings) 40) #27 or #28 or #29 or #30 or #31 or #32 or #33 or #34 or # 35 or #36 or #37 or #38 or #39. d) FILTER FOR HUMAN STUDIES 41) animal in de 42) animal (EXPLODE MeSH, all subheadings) 43) nonhuman (EXPLODE MeSH, all subheadings) 44) animal-experiment (EXPLODE MeSH, all subheadings) 45)#41 or #42 or #43 or #44 46) human in de
47) human (EXPLODE MeSH, all subheadings) 48) #46 or #47 49) #48 and #45 50) #45 not #49 51) #40 not #50 e) FINAL COMBINATION 52) #18 and #26 and #51 2. Handsearches The British Library conference proceedings index and the ISI proceeding of the Web of Knowledge were searched for conferences specic toPS. 3. Personal communication We attempted to contact experts in the eld to identify any further trial evaluations that were missed in the main search. 4. Reference lists Reference lists and forward citations of included trials and relevant systematic reviews were searched.
Data collection and analysis

Study selection In accordance with the dened inclusion criteria, citations were screened by two of the review authors. Where it was not possible to accept or reject, the full text of the citation was obtained for further evaluation. Following screening, the full text of eligible citations was assessed for inclusion independently by two of the review authors (BC and LJ). Where differences of opinion existed they were resolved by consensus with the other review authors (RW, MK and AT). If resolution was not possible, we attempted to contact the authors for clarication. Justication for excluding studies at this stage was documented. Data extraction and management Data was extracted independently by two review authors (BC and LJ). Where there are any discrepancies, the other review authors (RW, MK and AT) was consulted and discrepancy resolved by consensus A data extraction form was designed for this review. Where possible, the following information was obtained for each study: The number of patients eligible, number randomised, and reasons why patients not included in trial. The number of patients evaluated at follow-up(s) and what were the follow-up time points. Patient characteristics including age, sex, co-morbidities, diagnosis of depression. Trial design features on masking, whether parallel or crossover, features of randomisation, and sample size calculation. PS intervention including dosage, duration and mode. Comparison intervention including dosage, duration and mode. Outcome data at all time points including how it was measured, and the mean or categorical scores of the main and
5
other outcomes. Data from cross-over trials were only included if the researchers presented separately pre-cross over results or the means and standard deviations reported for each trial arm separately. Where information was lacking, we attempted to make contact with trial authors or trial sponsors. Assessment of trial quality Two review authors (BC and LJ) independently assessed the quality of included trials according to criteria described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2006) and supported by Juni 2001. Four main sources of systematic bias were assessed for each included trial: a) selection bias (randomisation sequence and concealment of allocation, and bias at recruitment) b) performance bias c) detection bias d) attrition bias (the completeness of follow up, with less than 10% loss to follow up dened as adequate) Criteria a) and d) was assessed as adequate, inadequate or unclear, according to Juni 2001. Criteria b) and c) was assessed by the use of three questions: a) were the trial participants aware of their assigned intervention (performance bias)? b) were those providing the intervention aware of the assigned intervention (performance bias)? c) were those who were assessing outcomes aware of the assigned intervention (detection bias)? Based on the quality criteria, if was planned that trials would be broadly subdivided into the following three categories: a) all quality criteria met: low risk of bias b) one or more of the quality criteria only partly met: moderate risk of bias and c) one or more criteria not met: high risk of bias. Measures of treatment of effect Trials measured treatment effect using either dichotomous data or a psychiatric ordinal rating scale for depression. Dichotomous data Where dichotomous data was reported, odds ratios (ORs) and their 95% condence intervals (CI) were generated. Continuous data Effects measures for ordinal data were assessed as continuous data. The weighted mean difference (WMD) was generated for ordinal data where the data was provided as a mean and standard deviation. Where different scales were used to pool data to measure the same outcome, the standardised mean difference (WMD) was used, together with 95% condence intervals (CI). Dealing with missing data Where data were missing we attempted to contact trial authors. For trials using continuous data in which SDs were not reported, and no information was available from the authors, the SD were calculated via the standard error of the mean (SEM) by multiplying
the SEM by the square root of the sample size. In the absence of a SEM it was planned that the SD would be derived from the pvalue using the method appropriate to the test conducted. When either of these was undertaken we planned to explore the effects in a sensitivity analysis. Drop-outs were included in analysis. Responders and remitters to treatment were calculated on the intention-to-treat (ITT) basis. Where participants had withdrawn from the trial before the endpoint, it was assumed they would have experienced the negative outcome by the end of the trial (e.g. failure to respond to treatment). When there was missing data and the method of last observation carried forward (LOCF) had been used to do an ITT analysis, then the LOCF data was used, with due consideration of the potential bias and uncertainty introduced. Data synthesis (meta-analysis) Where trial data were of sufcient quality and sufciently similar (in diagnostic criteria, intervention, outcome measure, length of follow-up and type of analysis) they were combined in a metaanalysis to provide a pooled effect estimate. A xed-effects model was used in the rst instance. If there was no statistical heterogeneity, a random-effects model was used to check the robustness of the xed-effect model. If statistical heterogeneity was observed, the random-effects model was used a priori. Assessment of heterogeneity Where meta-analysis was possible, statistical heterogeneity was assessed between trials using the chi-squared statistic and I-squared statistic (A chi-squared p-value of less than 0.05 or an I-squared value equal to or more than 50% was considered indicative of substantial variability in effect estimate between trials due to heterogeneity). If heterogeneity was identied subgroup analysis was undertaken to explore the lack of homogeneity. Subgroup analysis and investigation of heterogeneity To explore clinical heterogeneity and investigate the effect modication of participants and treatment types, we planned, if sufcient data were available, to perform the following subgroup analyses: Participants: (1) Gender - male versus female (2) Mild or moderate versus severe depression (according to clinical status) (3) Depression in medically ill patients Intervention : (3) Type of PS. Sensitivity analysis We planned to perform a sensitivity analyses in order to explore the inuence of the following factors: 1. Excluding unpublished studies (if there were any); 2. Taking account of study quality (low-, moderate- or high-risk of bias); 3. Excluding studies using the following lters: (a) diagnostic criteria used for depression
6
(b) source of funding (industry versus other) (c) scales used for measuring effect (validated versus other) Publication bias We planned to explore publication bias by using funnel plots.
RESULTS
Description of studies
See: Characteristics of included studies; Characteristics of excluded studies. Results of the search A total of 77 studies were initially identied through the CCDANCTR-Studies and CCDANCTR-References searches. A further 21 references were identied by the additional MEDLINE searches. Included studies Twenty-four RCTs published in twenty-three papers were identied that assessed the effectiveness of a PS in the treatment of the symptoms of depression (Cantello 1989; Cookson 1986; DeBattista 2003; Elizur 1979; Fava 2005; Fernandez 1995; Hare 1962; Hare 1964; Kits 1970; Lee 2005; Mattes 1985; Overall 1962; Patkar 2006; Postolache 1999; Reus 1979; Rickels 1970; Rickels 1972; Robin 1958; Silberman 1981; Vaishnavi 2006; Wagner 2000; Wallace 1995; Wheatley 1969 a; Wheatley 1969 b). The Characteristics of Included Studies table provides details of the included trials in terms of the populations studied, the treatments examined, and the outcome measures used. Key study characteristics are summarised below. Study design Fifteen of the 24 included trials were of parallel design. The others were of cross-over design. Fifteen trials were performed over 20 years ago. Four trials declared pharmaceutical funding or interests (DeBattista 2003; Fava 2005; Patkar 2006; Vaishnavi 2006). Sample sizes Most trials (18/24) involved less than 100 participants. In nine trials there were 10 or fewer participants per trial arm. Participants Just over half of the trials (14/24) were performed on North American populations. Six were performed on UK populations. Just over half of the trials (13/24) established baseline depression using the validated diagnostic schedule in the Diagnostic Schedule Manual (DSM) or the Research Diagnostic Criteria (RDC). The other trials relied on a clinical diagnosis. Twelve trials evaluated the effects of a PS on patients with major depression (Cantello 1989; DeBattista 2003; Elizur 1979; Fava 2005; Fernandez 1995; Lee 2005; Patkar 2006; Postolache 1999; Reus 1979; Silberman 1981; Vaishnavi 2006; Wallace 1995). Other trials evaluated effects in patients with mild depression (Mattes 1985;
Rickels 1972), or included patients whose depression ranged from mild to moderate (Rickels 1970; Robin 1958) or mild to severe (Wagner 2000). In seven other trials the severity of the depression was unclear (Cookson 1986; Hare 1962; Hare 1964; Kits 1970; Overall 1962; Wheatley 1969 a; Wheatley 1969 b). In ve trials patients were described as having a partial or non-response to antidepressant therapy (DeBattista 2003; Fava 2005; Patkar 2006; Vaishnavi 2006; Wallace 1995). Six trials included as part of their eligibility criteria other common symptoms of depression including fatigue reduced energy, hypersomnia and appetite changes (Elizur 1979; Fava 2005; Rickels 1970; Rickels 1972; Robin 1958; Wagner 2000). Three trials specied that they excluded patients with a history of drug abuse (Elizur 1979; Silberman 1981; Wagner 2000). In ve trials the participants had serious concomitant medical conditions, including Parkinsons disease (Cantello 1989), HIV ( Fernandez 1995; Wagner 2000), traumatic brain injury (Lee 2005) and chronic, debilitating physical illnesses (unspecied) in older patients (Wallace 1995). In most trials participants were adults aged less than 70 years. Four trials included people over 70 years of age (Cookson 1986; Rickels 1970; Wallace 1995; Wheatley 1969 b). Five studies did not state the maximum age of participants (Hare 1962; Hare 1964; Overall 1962; Rickels 1972; Robin 1958). With the exception of two trials that explored the effects of a PS on depression in men with HIV (Fernandez 1995 and Wagner 2000), all trials included male and female participants. Gender was not reported in one trial (Reus 1979). Interventions Just over half of the trials, 14/24, evaluated PS using an oral preparation as a monotherapy and a placebo control (Elizur 1979; Hare 1962; Kits 1970; Lee 2005; Mattes 1985; Overall 1962; Rickels 1970; Rickels 1972; Robin 1958; Vaishnavi 2006; Wagner 2000; Wallace 1995; Wheatley 1969 a; Wheatley 1969 b). Six trials evaluated a PS either as an adjunct or where patients continued on an antidepressant (DeBattista 2003; Fava 2005; Patkar 2006; Postolache 1999) or a barbiturate (Hare 1964; Overall 1962). Four trials evaluated a PS as an intravenous monotherapy (Cantello 1989; Cookson 1986; Reus 1979; Silberman 1981). Treatment duration ranged from a single dose (Cantello 1989; Cookson 1986; Reus 1979; Silberman 1981) to 12 weeks therapy (Overall 1962; Vaishnavi 2006). In the parallel design trials most treatments were of 3-4 weeks duration. In the cross-over trials duration per treatment was commonly two weeks or less. All intravenous therapy interventions consisted of a single injection. A high proportion of evaluations (10/24) were of the effect of the PS methylphenidate (Cantello 1989; Fernandez 1995; Lee 2005; Mattes 1985; Patkar 2006; Postolache 1999; Rickels 1970; Rickels 1972; Robin 1958; Wallace 1995). Four other types of PS were evaluated: 1. Dexamphetamine (Hare 1962; Hare 1964; Overall 1962; Reus 1979; Silberman 1981; Wagner 2000; Wheatley 1969 a; Wheatley
7
1969 b) 2. Methylamphetamine (Cookson 1986; Kits 1970) 3. Modanil (DeBattista 2003; Fava 2005; Vaishnavi 2006) 4. Pemoline (Elizur 1979; Rickels 1970) Outcomes The effects of a PS on depressed mood were commonly evaluated using a version of the Hamilton Depression Scale (HAM-D). Other symptoms associated with depression such as hypersomnia, fatigue or reduced energy were measured either by a validated psychometric scales, including Epworth Sleepiness Scale and the Brief Fatigue Inventory, or by self-report. Less than half of the trials, 9/24, set out a priori to measure side effects (Cantello 1989; DeBattista 2003; Elizur 1979; Fava 2005; Lee 2005; Mattes 1985; Overall 1962; Patkar 2006; Vaishnavi 2006). Four trials did not report reasons for participants dropping out (Lee 2005; Mattes 1985; Patkar 2006; Robin 1958). Excluded studies Seventy-seven trials were excluded commonly because they did not evaluate a PS or were not RCTs. The reasons for exclusion are listed in the Characteristics of excluded studies table.
Risk of bias in included studies

The overall quality of the trials was poor. Many trials pre-dated the Consort statement, and as they were undertaken at least 20 years ago, it was not possible to contact authors. All trials appeared to be vulnerable to a number of biases that may have affected their results. Selection bias Very few trials (four out of the 24 included trials) provided sufcient details on randomisation generation. One trial provided some details on concealment allocation. None of the trials provided details on potential bias at recruitment. Detection and performance bias Although all trials were described as double blind, few (4/24) provided any details on who was masked (Hare 1964; Robin 1958; Reus 1979; Wallace 1995).. Attrition bias In ten trials there was some risk of attrition bias, as more than 10% of participants were lost to follow-up. Only one of these trials used intention to treat analysis (Patkar 2006). See Additional Table 1 for further details.
Effects of interventions
Thirteen trials had data in a suitable form for meta-analysis ( DeBattista 2003; Elizur 1979; Fava 2005; Fernandez 1995; Lee 2005; Patkar 2006; Postolache 1999; Rickels 1970; Rickels 1972; Robin 1958; Vaishnavi 2006; Wagner 2000; Wheatley 1969 a). Trials with unusable data for meta-analysis included all nine crossover trials, as none provided separately pre-crossover results, or
provided a mean difference (with standard deviation) between trial arms. Analyses were presented according to short-term PS use (trial duration of 4 weeks or less) (Elizur 1979; Lee 2005; Patkar 2006; Rickels 1970; Rickels 1972; Robin 1958; Wagner 2000; Wheatley 1969 a) or medium term use (trial duration of 5-12 weeks) (DeBattista 2003; Fava 2005; Fernandez 1995;Postolache 1999; Vaishnavi 2006). Modanil was examined separately from other psychostimulants in all meta-analyses, due to its different mode of action. All trials included in the meta-analyses used an oral therapy route. The xed effect model (FE) was used in the rst instance in analyses, with the random effects model (RE) used to test the robustness of the ndings. Unless otherwise stated below, the xed effects model was used in analyses. Where statistical heterogeneity was observed, the random effects model was used. COMPARISON 1: PSYCHOSTIMULANTS VERSUS PLACEBO (Graphs 01 01 to 01 17) Eight trials contributed to this comparison (Elizur 1979; Lee 2005; Rickels 1970; Rickels 1972; Robin 1958; Vaishnavi 2006;Wagner 2000; Wheatley 1969 a) Primary outcome Reduction in depression symptoms (Graph 01 01 to 01 02) Three short term trials of PS, with a total of 62 participants, were included in the meta-analysis. All trials used a validated depression scale. There was a signicant difference in favour of PS (SMD 0.87, 95% CI -1.40, -0.33). The I2 test at 15.5% did not indicate a high degree of heterogeneity. The random effects model did not differ greatly from the xed effect model. In one medium term trial of Modanil, with a total of 50 participants, there was no signicant difference in depression scores between Modanil and placebo (WMD -2.70, 95% CI -7.76, 2.36). Clinical response (Graph 01 03) Three short term trials of PS (130 participants) measured improvement in depression symptoms, two by clinical assessment and one using a validated scale. There was no signicant difference in improvement in symptoms (OR 1.01, 95% CI 0.48, 2.09). The I2 test at 23.1% did not indicate a high degree of heterogeneity. The random effects model did not differ greatly from the xed effect model. One trial measured clinical response as 50% reduction in depression scale. No signicant difference was identied between PS and placebo (OR 1.83, 95% CI 0.57, 5.88). Secondary outcomes 1. Changes in other symptoms associated with depression Cognitive functioning (Graph 01 04) Two short term trials of PS, with a total of 42 participants, were included in this analysis, with cognitive functioning measured by Mini-Mental State Examination and Trail Making Test A. No overall signicant difference was observed between PS and control (SMD 0.42, 95% CI -0.20, 1.04). The I2 test at 16% indicated a low degree of heterogeneity. The random effects model did not differ from the xed effects model.
8
Fatigue (Graph 01 05 to 01 06) Of the two short term trials of PS recording fatigue scores, there was a signicant difference between the PS and placebo groups that favoured PS (SMD -1.80, 95% CI -2.54, -1.06). The trials were homogenous with an I2 of 0%. The random effects model did not differ from the xed effects model. There was no signicant difference in fatigue scores between modanil and placebo in the one medium term trial included in this analysis (WMD 0.24, 95% CI -1.44,1.92.) Hypersomnia (Graph 01 07) In one medium term trial of PS that recorded hypersomnia mean scores, there was no signicant difference between the PS and placebo groups (WMD -5.00, 95% CI -10.87, 0.87). In one medium term trial of modanil that recorded hypersomnia mean scores, using the Epworth Sleepiness Scale, there was no signicant difference between the modanil and placebo groups (WMD -0.24, 95% CI -3.06, 2.58). Other symptoms associated with depression See Additional Table 2. There were no data on other symptoms associated with depression including: changes in weight when not dieting, weight gain, changes in appetite; psychomotor agitation or retardation; recurrent thoughts of death or suicidal ideation. 2. Remission No studies were included in this outcome 3. Social adjustment and functioning In one trial of PS with a total of 22 participants, there was no signicant difference in the short term on physical functioning (WMD 10.90, 95% CI 0.67, 21.13). No trials evaluated social adjustment or social functioning. 4. Health-related quality of life (Graph 01 08) Two short term trials of PS with a total of 42 participants were included in the analysis. No signicant overall effect was observed in mean at follow up between PS and control (SMD 0.57, 95% CI -0.05, 1.20). The I2 test at 0% did not indicate heterogeneity. The random effects model did not greatly differ from the xed effects model. 5. Acceptability Dropouts due to side effects (Graph 01 09) Two short term trials of PS with a total of 68 participants were included in this analysis of short term effect. No overall signicant difference was observed between PS and control (OR 1.63, 95% CI 0.26, 10.25). The I2 test at 34.3% did not indicate signicant heterogeneity. The random effects model did not greatly differ from the xed effect model. One small medium term trial with a sample of 9 showed no signicant difference between the two groups for dropout rates due to side effects (OR 6.43, 95% CI 0.23, 181.82). Dropouts due to inefcacy No studies included these data
6. Tolerability Proportion of participants experiencing some side effects (Graph 01 10 to 01 11) Two short term trials of PS with a total of 185 participants were included in this analysis. No overall signicant difference was observed between PS and control, using a random effects model (OR 1.78, 95% CI 0.57, 5.55). The I2 test at 46% indicated some heterogeneity. The random effects model did not greatly differ from the xed effects model. Two medium term trials of PS with a total of 90 participants were included in this analysis. A signicant difference favouring the control group was observed between PS and control, using a random effects model (OR 7.22, 95% CI 2.21, 23.57). Dizziness (Graph 01 12) Two short term trials of PS with a total of 190 participants were included in this analysis. No overall signicant difference was observed between PS and placebo control (OR 1.31, 95% CI 0.40, 4.24). The I2 test at 0% did not indicate heterogeneity. The random effects model did not greatly differ from the xed effects model. Drowsiness (Graph 01 13) Two short term trials of PS with a total of 190 participants were included in this analysis. No overall signicant difference was observed between PS and placebo control (OR 0.41, 95% CI 0.12, 1.37). The I2 test at 0% did not indicate heterogeneity. The random effects model did not greatly differ from the xed effects model. Headache (Graph 01 14) Two short term trials of PS with a total of 115 participants were included in this analysis. No signicant difference was observed between PS and placebo control (OR 1.06, 95% CI 0.34, 3.35). The I2 test at 0% did not indicate heterogeneity. The random effects model did not greatly differ from the xed effects model. Insomnia (Graph 01 15) One short term trial of PS with 95 participants was included in this analysis. No signicant difference was observed between PS and placebo control (OR 1.89, 95% CI 0.66, 5.38). The I2 test at 0% did not indicate heterogeneity. The random effects model did not greatly differ from the xed effects model. Nausea (Graph 01 16) Two short term trials of PS with a total of 190 participants were included in this analysis. No signicant difference was observed between treatments (OR 1.52, 95% CI 0.31, 7.57). The I2 test at 0% did not indicate heterogeneity. The random effects model did not greatly differ from that in the xed effects model. Tremor (Graph 01 17) Two short term trials of PS with a total of 115 participants showed no signicant difference between PS and placebo control (OR 2.83 95% CI 0.51, 15.67). The I2 test at 0% did not indicate heterogeneity. The random effects model did not greatly differ
9
from the xed effects model. Assessment of other side effects No trials evaluated PS misuse and dependence, withdrawal reactions, over arousal, suicidal thoughts and acts of self-harm. COMPARISON 2: PSYCHOSTIMULANTS VERSUS ANTIDEPRESSANTS (Graphs 02 01 to 02 02) Two trials contributed to this comparison (Lee 2005; Fernandez 1995). Primary outcome Reduction in depression symptoms (Graph 02 01) In one small short term trial of PS with a total of 20 participants, there was no signicant difference in depression scores between the PS and antidepressant groups (WMD -4.30, 95% CI --8.79, 9.19). Secondary outcomes Tolerability Proportion of participants experiencing side effects (Graph 02 02) In one small medium term trial of PS with a total of 15 participants, there was no signicant difference between the PS and antidepressant groups in the number of participants experiencing side effects (OR 1.43, 95% CI 0.10, 20.44) Headache In one trial there was no signicant difference between treatment arms (OR 0.23, 95% CI 0.01, 5.73). Insomnia In one trial there was no signicant difference between treatment arms (OR 0.25, 95% CI 0.02, 3.10). Nausea In one trial there was no signicant difference between treatment arms (OR 4.00, 95% CI 0.27, 58.56). Other secondary outcomes No other secondary outcomes were included in this comparison. COMPARISON 3: PSYCHOSTIMULANTS VERSUS PLACEBO AS ADJUNCT TO ANTIDEPRESSANT TREATMENT Four trials contributed to this comparison (DeBattista 2003 ; Fava 2005 ; Patkar 2006;Postolache 1999) Primary outcome Reduction in depression symptoms (Graph 03 01 to 03 02) In one short term trial of PS with a total of 50 participants, there was no signicant difference in the mean depression score between PS versus placebo as adjunct to antidepressant treatment (WMD -1.60, 95% CI -5.96, 2.76). In two short term trials of modanil with a total of 411 participants, there was no statistical difference in depression scores between modanil and placebo (WMD -0.09, 95% CI -1.07, 0.89). The I2 test at 0% did not indicate heterogeneity. The random effects model did not greatly differ from the xed effects model. In two medium term trials of modanil, with a total of 443 participants, there was no signicant difference in mean depression
scores between modanil and placebo (WMD -0.80, 95% CI1.88, 0.28). The I2 test at 48.5% indicated a moderate degree of heterogeneity. The random effects model did not greatly differ from the xed effects model. Clinical response (Graph 03 03) In one short term trial of PS with a total of 60 participants, there was a non-signicant difference in 50% response rate (OR 1.83, 95% CI 0.57, 5.88) and in clinical improvement (OR 2.19, 95% CI 0.72, 2.34) between PS and placebo. In one medium term trial of PS with a total of 9 participants, there was no signicant difference in clinical improvement between PS combined with adjunct AD treatment versus placebo with AD treatment (OR 1.10, 95% CI 0.31, 3.91). Secondary outcomes 1. Changes in other symptoms associated with depression Fatigue (Graph 03 04 to 03 05) Of the two short term trials of modanil recording mean fatigue scores, there was no signicant difference between treatment and control, using a random effects model (WMD -0.04, 95% CI 0.79, 0.87). The I2 test at 82.3% indicated a high degree of heterogeneity. Of the two medium term trials of modanil recording mean fatigue scores, there was no signicant difference between treatment and control (WMD -0.31, 95% CI -0.67, 0.05). The I 2 test at 44.1% indicated a moderate degree of heterogeneity. Hypersomnia (Graph 03 06 to 03 07) Of the two short term trials of modanil recording mean hypersomnia scores, there was no signicant difference between treatment and control (WMD -0.79, 95% CI -1.63, 0.06). The I2 test at 0% did not indicate heterogeneity. The random effects model did not differ from the xed effects model. Of the two medium term trials of modanil recording mean hypersomnia scores, there was a signicant difference between treatment and control (WMD -1.27, 95% CI -2.14, -039). I2 test at 0% did not indicate heterogeneity. The random effects model did not differ from the xed effects model. Other symptoms associated with depression See Additional Table 2. There were no data on other symptoms associated with depression including: changes in weight when not dieting, weight gain, changes in appetite; psychomotor agitation or retardation; recurrent thoughts of death and suicidal ideation. 2. Remission No studies were included in this outcome 3. Social adjustment and functioning No studies were included in this outcome 4. Health-related quality of life No studies were included in this outcome 5. Acceptability Dropouts due to side effects (Graph 03 08 to 03 09) In one small medium term trial of PS with a total of 9 participants, there was no signicant difference in dropout rates due to side
10
effects between the PS and placebo groups (OR 6.43, 95% CI 0.23, 181.82). Two medium term trials of modanil with a total of 450 participants were included in this analysis. No overall signicant difference was observed between modanil and placebo (OR 1.31, 95% CI 0.54, 3.18). The I2 test at 0% did not indicate heterogeneity. The random effects model did not greatly differ from the xed effect model. Dropouts due to lack of efcacy (Graph 03 10) Two medium term trials of modanil with a total of 450 participants were included in analysis of the proportion of participants who dropped out due to inefcacy. No signicant difference was observed between Modanil and placebo control (OR 0.97, 95% CI 0.24, 3.96). The I2 test at 0% did not indicate heterogeneity. The random effects model did not greatly differ from the xed effect model. 6. Tolerability Proportion of participants experiencing side effects (Graph 03 11) One short term trial of PS comparing PS as an adjunct to ADs against placebo and ADs showed a non-signicant difference in the proportion of participants experiencing side effects (OR 1.32, 95% CI 0.47, 3.72) Headache (Graph 03 12) Two medium term trials of modanil, with a total of 447 participants, showed a lack of signicant difference in the proportion of participants experiencing headaches between the two groups, using a random effects model (OR 1.21, 95% CI 0.50, 2.94). The I2 test at 60% indicated a substantial degree of heterogeneity. Insomnia (Graph 03 13 to 03 14) One short term trial of PS, involving 60 participants, showed a lack of signicant difference in the proportion of participants experiencing insomnia between the two groups (OR 1.00, 95% CI 0.19, 5.40). Two medium term trials of modanil, with a total of 447 participants, showed a lack of signicant difference in the proportion of participants experiencing insomnia between the two groups (OR 1.24, 95% CI 0.62, 2.50). The I2 test at 0% did not indicate heterogeneity. Nausea (Graph 03 15 to 03 16) One short term trial of PS, with 60 participants, showed no significant difference between treatment groups for nausea (OR 3.22, 95% CI 0.62, 2.50). Two medium term trials of modanil, with a total of 447 participants, showed no signicant difference between the two groups, using a random effects model (OR 1.78, 95% CI 0.20, 15.95). The I2 test at 80% indicated a high degree of heterogeneity. Tremor (Graph 03 17) One short term trial of PS, with 60 participants, showed no significant difference between treatment groups for tremor (OR 0.48,
95% CI 0.04, 5.63). Upper respiratory tract infection (Graph 03 18) Two medium term trials, with a total of 447 participants, were included in this analysis. No overall signicant difference was observed between PS and placebo control (OR 0.73, 95% CI 0.31, 1.71). The I2 test at 0% did not indicate heterogeneity. The random effects model did not greatly differ from the xed effects model. Assessment of other side effects No trials evaluated PS misuse and dependence, withdrawal reactions, over arousal, suicidal thoughts and acts of self-harm. Other side effects reported in one trial No signicant difference was observed between PS and control treatment in asthenia, cardiovascular vital signs, constipation, excitement, hypertension, hypertonia, infection, loss of appetite, myalgia, nasopharngitis, nervousness, palpitation, restlessness, rhinitis, tachycardia, tolerance, vertigo, and weight changes. See Additional Table 3 for further details. Severity of adverse events None of the side effects were reported as being moderate or severe, bar one person on PS in one trial (Patkar 2006), who had an episode of non-cardiac chest pain. Subgroup analysis Depression in medically ill patients (Graph 04 01) In two short term trials of PS with a total of 42 participants, there was a signicant difference favouring PS compared with placebo (SMD -0.75, 95% CI -1.39, -0.11). The I2 test at 48.5% indicated a moderate degree of heterogeneity. The random effects model did not differ greatly from the xed effects model but it was no longer signicant. Because of insufcient trial data, it was not possible to undertake subgroup analyses of gender, severity of depression or type of PS . Sensitivity analysis It was not possible to undertake any sensitivity analyses of poorer quality trials or those that used non-validated measures because of the small number of trials included in the review . Secondly, we could not exclude unpublished trials, as all trials were published. Thirdly, it was not possible to exclude industry funded trials, because all data in meta-analyses were either all non-industry funded or, in the case of meta-analyses of trials that used Modanil, all were industry funded. Evidence from trials not included in any meta-analysis of effect of PS on depression Because of limitations in reporting of data, 13 trials could not be included in any of the meta-analyses examining the effect of PS on depression symptoms. Nine of these trials individually found no signicant difference in effect between PS and control treatment in reducing depression (See Additional Table 4). Publication bias The presence of publication bias was not examined in this systematic review, as too few trials with suitable data were found to
11
construct a meaningful funnel plot.
arousal. In addition, although depression is associated with suicide and deliberate self-harm, these outcomes were not reported. Potential biases in the review process
DISCUSSION
Summary of main results The purpose of this systematic review was to conduct an evaluation of the available evidence regarding the effectiveness and safety of psychostimulants (PS) for depression. As a result of heterogeneity in PS intervention and comparative treatments, and the paucity of RCTs with sufcient data for qualitative analysis, few clinically relevant conclusions can be drawn. The meta-analyses of the primary outcome included an analysis of three trials, involving 62 participants, and pooled scores from a validated depression scale. The results obtained using xed effects models suggest that for people with depression, treatment with oral PS in comparison with a placebo in the short term (up to four weeks) signicantly reduces symptoms. Statistical tests did not indicate a high degree of heterogeneity and use of the more conservative random effects statistical model did not change the results. This effect was not replicated in meta-analysis of trials that measured effectiveness by improvement using a dichotomous measure, however not all trials (2/3) used a validated measure to evaluate improvement in depression. The effect was also not replicated in the meta-analysis of trials that used modanil. The robustness of our ndings of the positive effect of PS on depression in the short term could only be explored in one of the four a priori sensitivity and subgroup analyses we planned to undertake. This analysis conrmed the overall ndings. Pooling of outcomes for other symptoms of depression suggested that PS as a monotherapy has a favourable effect in the short term on fatigue. PS were found to be, in the short term, welltolerated and acceptable to patients. We are unable to comment in this review on the use of modanil for the relief of somnolence induced by opioid therapy or its effectiveness in the treatment of narcolepsy. There was insufcient evidence for us to make any comment about the potential for dependence on PS. Quality of the evidence There are limitations to our review ndings. Because of insufcient data, few trials were combined within a meta-analysis. Many of the trials lacked important information on quality, in particular on randomisation procedures. Most trials were relatively small. Apart from all the modanil trials, few of the trials with analyzable data exceeded 25 patients per group. Furthermore, it cannot be ruled out that there were differences in the diagnostic assessment of the patients and thus in the quality of the diagnosis across the trials. Despite these being potential reasons why PS are little used, no trials evaluated misuse, dependence, withdrawal reactions, and over
As with all systematic reviews, publication bias is a potentially serious source of bias, and in this review there were too few studies to investigate the possibility of publication bias using funnel plots.
AUTHORS CONCLUSIONS Implications for practice

There is insufcient evidence for this review to recommend the use of psychostimulants (PS) above other more established treatments of depression. Evidence to date suggests clinicians may consider PS appropriate in certain circumstances such as where established treatments for depression have failed and where treatment is required in the short term. The evidence to date does not support the use of modanil in the treatment of depression. Clinicians should note that although this review did not identify any serious adverse effect, pemoline (which was used in one of the included trials) is no longer licensed in UK and USA because of risk of liver toxicity.
Implications for research

Larger high quality randomised controlled trials with follow-up periods greater than four weeks are needed. Such trials should test the robustness of our ndings on short-term effects and explore the effect in certain subgroups of depressed patients, in particular, those with serious concomitant diseases such as patients receiving palliative care. Such trials should consider evaluating the effect of PS on all common depressive symptoms, including both further evaluations of the effect on fatigue and hypersomnia and further exploration of acceptability, tolerability and the potential for dependence and abuse.
ACKNOWLEDGEMENTS
The authors thank Dr M.E. Thase, Dr C. Pae, Dr K.Peindl, Dr K.Gadde and Professor J.Yoon for giving additional information about their trials. We would also like to thank representatives of Norvaris, Cephalon and Janssen-Cilag for their assistance in providing further information on published trials and/or in searching for any relevant unpublished or in progress trials. We also gratefully acknowledge project funding from Marie Curie Cancer Care, Dr C Stanford for pharmacological advice and the help of Hugh McGuire of the Cochrane Collaboration Depression, Anxiety and Neurosis Group in conducting the database searches.
12
REFERENCES
References to studies included in this review

Cantello 1989 {published data only} Cantello R, Aguggia M, Gilli M, Delsedime M, Chiardo Cutin I, Riccio A, et al.Major depression in Parkinsons disease and the mood response to intravenous methylphenidate: Possible role of the hedonic dopamine synapse. Journal of Neurology, Neurosurgery and Psychiatry 1989;52:72431. Cookson 1986 {published data only} Cookson J, Silverstone T. The effects of Methylamphetamine on Mood and Apetite in Depressed patients: A placebocontrolled study. International Clinical Psychopharmacology 1986;1:12733. DeBattista 2003 {published data only} DeBattista C, Doghramji K, Menza MA, Rosenthal MH, Fieve RR. Adjunct modanil for the short-term treatment of fatigue and sleepiness in patients with major depressive disorder: a preliminary double-blind, placebo-controlled study. Journal of Clinical Psychiatry 2003;64:105764. Elizur 1979 {published data only} Elizur A, Wintner I, Davidson S. The clinical and psychological effects of pemoline in depressed patients. A controlled study. International Pharmacopsychiatry 1979;14: 12734. Fava 2005 {published data only} Fava M, Thase ME, DeBattista C. A multicenter, placebocontrolled study of modanil augmentation in partial responders to selective serotonin reuptake inhibitors with persistent fatigue and sleepiness. Journal of clinical psychiatry 2005;66:8593. Fernandez 1995 {published data only} Fernandez F, Levy JK, Samley RH. Effects of methylphenidate in HIV-related depression: A comparative trial with desipramine. International Journal of Psychaitry in Medicine 1995;25:5367. Hare 1962 {published data only} Hare EH, Dominian J, Sharpe L. Phenelzine and dexamphetamine in depressive illness: A comparative trial. British Medical Journal 1962;1:912. Hare 1964 {published data only} Hare EH, McCance C, McCormick WO. Imipramine and Drinamyl in depressive illness: A comparative trial. British Medical Journal 1964;1:81820. Kits 1970 {published data only} Kits TP, van Praag. A controlled study of the antidepressant effect of P-chloro-n-methylamphetamine, a compound with a selective effect on the central 5-hydroxytryptamine metabolism. Acta Psychiatrica Scandinavica 1970;46: 36573. Lee 2005 {published data only} Lee H, Kim SW, Kim JM, Shin IS, Yang SY, Yoon JS. Comparing effects of methylphenidate, sertraline and placebo on neuropsychiatric sequelae in patients with
traumatic brain injury. Human Psychopharmacology 2005; 20:97104. Mattes 1985 {published data only} Mattes JA. Methylphenidate in mild depression: A doubleblind controlled trial. Journal of Clinical Psychiatry 1985; 46:52557. Overall 1962 {published data only} Overall JE, Hollister LE, Pokorny AD, Casey JF, Katz G. Drug therapy in depressions: Controlled evaluation of imipramine, isocarboxide, dextroamphetamine-amobarbital and placebo. Clinical Pharmacoly Therapy 1962;3:1622. Patkar 2006 {published data only} Patkar AA, Massand PS, Pae CU, Peindl K, HooperWood C, Mannelli P, et al.A randomized, double-blind, placebo-controlled trial of augnebtation with an extended release formulation of methylphenidate in outpatients with treatment resistant depression. Journal of Clinical Psychopharmacology 2006;26:6536. Postolache 1999 {published data only} Postolache TT, Rosenthal RN, Hellerstein DJ, Aromin R, Kelton GM, Muran JC, Londono JH. Early augmentation of sertraline with methylphenidate.. Journal of Clinical Psychiatry 1999;2:1234.. Reus 1979 {published data only} Reus VI, Silberman E, Post RM, Weingartner H. dAmphetamine: effects on memory in a depressed population. Biological Psychiatry 1979;14:34556. Rickels 1970 {published data only} Rickels K, Gordon P, Gansman D, Charles W, PereiraOgan J, Hesbacher P. Pemoline and methylphenidate in mildly depressed outpatients. Clinical Pharmacology and Therapeutics 1970;11:698710. Rickels 1972 {published data only} Rickels K, Gingrich RL, McLaughlin FW, Morris RJ, Sablosky L, Silverman H, et al.Methylphenidate in mildly depressed outpatients. Clinical Pharmacology and Therapeutics 1972;13:595601. Robin 1958 {published data only} Robin AA, Wiseberg S. A controlled trial of methyl phenidate (Ritalin) in the treatment of depressive states. Journal of Neurology, Neurosurgery and Psychiatry 1958;21: 557. Silberman 1981 {published data only} Silberman EK, Reus VI, Jimerson DC, Lynott AM, Post RM. Heterogenity of Amphetamine response in depressed patients. American Journal of Psychiatry 1981;10:13027. Vaishnavi 2006 {published data only} Vaishnavi S, Gadde K, Alamy S, Zhang W, Connor K, Davidson JR. Modanil for atypicall depression: effects of open-label and double-blind discontinuation treatment. Journal of Clinical Psychopharmacology 2006;26:3738.
13
Wagner 2000 {published data only} Wagner GJ, Rabkin R. Effects of dextroamphetamine on depression and fatigue in men with HIV: a double-blind, placebo-controlled trial. Journal of Clinical Psychiatry 2000; 61:43640. Wallace 1995 {published data only} Wallace AE, Kofoed LL, West AN. Double-blind, placebocontrolled trial of methylphenidate in older, depressed, medically ill patients. American Journal of Psychiatry 1995; 6:92931. Wheatley 1969 a {published data only} Wheatley D. Amphetamines in general practice: their use in depression and anxiety.. Seminars in Psychiatry 1969;1: 16373. Wheatley 1969 b {published data only} Wheatley D. Amphetamines in general practice: their use in depression and anxiety. Seminars in Psychiatry 1969;1: 16373..
Capstick 1965 {published data only} Capstick N, Corbett M, PAre C, Linford Rees W. A comparative trial of diazepam (valium) and amylobarbitone. British Journal of Psychiatry 1965;111:5179. Carlson 1995 {published data only} Carlson GA, Rapport MD, Kelly KL, Pataki CS. Methylphenidate and desipramine in hospitalized children with comorbid behaviour and mood disorders: separate and combined effects on behavior and mood. Journal of Child and Adolescent Psychopharmacology 1995;5:191204. Carlson 2004 {published data only} Carlson PJ, Merlock MC, Suppes T. Adjunctive stimulant use in patients with bipolar disorder. Bipolar Disorders 2004;6:41620. Carroll 1995 {published data only} Carroll KM, Nich C, Rounsaville BJ. Differential symptom reduction in depressed cocaine abusers treated with psychotherapy and pharmacotherapy. Journal of Nervous and Mental Disease 1995;183:2519. Christensen 1990 {published data only} Christensen L. Dietary treatment of depression. Behavior Therapy 1990;21:18394. Ciraulo 2005 {published data only} Ciraulo DA, Knapp C, Rotosen J, Sarid-Segal O, Ciraulo AM, LoCastro J, et al.Nefazodone treatment of cocaine deprendence with comorbid depressive symptoms. Addiction 2005;100:2331. Clauson 2004 {published data only} Clauson AS, Elliott ES, Watson BD, Treacy J. Coadminstration of Phenelzine and Methylphenidate for treatment resistant depression. Annals of Pharmacotherpy 2004;38:508. Coffey 1990 {published data only} Coffey CE, Figiel GS, Weiner RD, Saunders WB. Caffeine augmentation of ECT. American Journal of Psychiatry 1990; 147:57985. Cornelius 1994 {published data only} Cornelius JC. Fluoxetine versus placebo in depressed alcoholics. Psychopharmacology 1994;30:661. Cornelius JR, Perkins KA, Salloum IM, Thase ME, Moss HB. Fluoxetine versus placebo to decrease the smoking of depressed alcoholic patients. Journal of Clinical Psychopharmacology 1999;19:1834. Cornelius JR, Salloum IM, Cornelius MD, Perel JM, Ehler JG, Jarrett PJ, et al.Preliminary report: doubleblind, placebo-controlled study of uoxetine in depressed alcoholics. Psychopharmacology Bulletin 1995;31:297303. Cornelius JR, Salloum IM, Ehler JG, Jarrett PJ, Cornelius MD, Black A, et al.Double-blind uoxetine in depressed alcoholic smokers.. Psychopharmacology Bulletin 1997;33: 16570. Cornelius JR, Salloum IM, Ehler JG, Jarrett PJ, Cornelius MD, Perel JM, et al.Fluoxetine in depressed alcoholics. A
14
References to studies excluded from this review

Ballinger 1974 {published data only} Ballinger BR, Presly A, Reid AH, Stevenson IH. The effects of hypnotics on imipramine treatment. Psychopharmacology 1974;39:26774. Blashki 1971 {published data only} Blashki TG, Mowbray R, Davies B. Controlled trial of amitriptyline in general practice. British Medical Journal 1971;1:1338. Blashki TG, Mowbray R, Davies BM. A controlled trial of an antidepressant (amitriptyline) in general practice. V World Congress of Psychiatry, Nov 28-Dec 4, Ciudad De Mexico. 1971. Brooks 2003 {published data only} Brooks A, Aharonovich E, Nunes E. Early participation in relapse prevention therapy predicts abstinence in a randomized clinical trial of depramine with depressed cocaine abusers. Annual Scientic Meeting of the College on Problems of Drug Dependence, Bal Harbour, FL, June 15-19. 2003. Buchsbaum 1977 {published data only} Buchsbaum MS, Van Kammen DP, Murphy DL. Individual differences in average evoked responses to d- and lamphetamine with and without lithium carbonate in depressed patients. Psychopharmacology 1977;51:12935. Buckman 1973 {published data only} Buckman J, Hain JD, Smith BM, Stevenson I. Controlled Interviews using drugs II. Comparisons between restricted and freer conditions. Archives of General Psychiatry 1973;29: 6237. Camacho 2006 {published data only} Camacho A, Ng B. Methylphenidate for alpha-interferon induced depression. Journal of Psychopharmacology 2006; 20:6879.
double-blind placebo-controlled trial. Archives of General Psychiatry 1997;54:7005. Cornelius JR, Salloum IM, Ehler JG, Jarrett PJ, Perel J, Thase ME. Double-blind uoxetine in depressed alcoholics. 149th Annual Meeting of the American Psychiatric Association; Nay 4-9; New York, NY. 1996. Cornelius JR, Salloum IM, Haskett RF, Daley DC, Cornelius MD, Thase ME, et al.Fluoxetine versus placebo in depressed alcoholics: A 1 year follow-up study. Addictive Behaviors 2000;25:30710. Salloum IM, Thase ME, Haskett RF, Daley DC, Jones Barlock A, Upsher C, et al.Fluoxetine versus placebo in depressed alcoholic cocaine abusers. Psychopharmacology Bulletin 1998;34:11721. Daley 1998 {published data only} Daley DC, Salloum IM, Zuckoff A, Kirisci L, Thase ME. Increasing treatment adherence among outpatients with depression and cocaine dependence: Results of a pilot study. American Journal of Psychiatry 1998;155:16113. Darvill 1959 {published data only} Darvill FT, Woolley S. Double-blind evaluation of methylphenidate (Ritalin) hydrochloride. JAMA 1959;169: 173941. de Wit 1987 {published data only} de Wit, Uhlenhuth EH, Johanson CE. The reinforcing properties of amphetamine in overweight subjects and subjects with depression. Clinical Pharmacology and Therapeutics 1987;42:12736. Duval 1996 {published data only} Duval F, Mokrani MC, Crocq MA, Jautz M, Bailey P, Diep TS, et al.Effect of anti-depressant medication on morning and evening thyroid function tests during a major depressive episode. Archives of General Psychiatry 1996;53:83340. Fava 2006 {published data only} Fava M, Rush AJ. Current status of augmentation and combination treatment for major depressive disorder: a literature review and a proposal for a novel approach to imporive practice. Psychotherpay and Psychosomatics 2006; 75:13953. Fawcett 1971 {published data only} Fawett J, Siomonpoulos V. Dextroamphetamine response as a possible predictor of improvement with tricyclic therapy in depression. Archives of General Psychiatry 1971;25:24755. Fawcett 1972 {published data only} Fawsett J, Maas JW, Dekirmenjian H. Depression and MHPG Excretion. Archives of General Psychiatry 1972;26: 24651. Fawcett 1987 {published data only} Fawcett J, Edwards JH, Kravitz HM, Jeffries H. Alprazolam: an antidepressant? Alprazolam, desipramine, and an alprazolam-desipramine combination in the treatment of adult depressed outpatients. Journal of Clinical Psychopharmacology 1987;7:295310. Kravitz HM, Edwards JH, Fawcett J, Fogg L. Challenging the amphetamine challenge test: Report of antidepressant
treatment study. Journal of Affective Disorders 1990;20: 1218. Giannini 1986 {published data only} Giannini AJ, Malone DA, Giannini MC, Price WA, Loiselle RH. Treatment of depression in chronic cocaine and phencyclidine abuse with desipramine. Journal of Clinical Pharmacology 1986;26:2114. Grade 1998 {published data only} Grade C, Redford B, Chrostowski J, Toussaint L, Blakewell B. Methylphenidate in early poststroke recovery: a doubleblind, placebo-controlled study. Archives of Physical Medicine and Rehabilitation 1998;79:104750. Harris 1960 {published data only} Harris JA, Robin AA. A controlled trial of Phenelzine in depressive reactions. Journal of Mental Science 1960;106: 14327. Hicks 1994 {published data only} Hicks P, Matthews TK, Riggs M. Caffeine and ECT in the treatment of depression. Psychopharmacology Bulletin 1994; 30(1):106. Hinton 1959 {published data only} Hinton JM. A comparison of Perphenazine, (Fentazin) Sodium Amylobarbitone and a placebo in anxious and depressed out-patients. British Journal of Psychiatry 1959; 105:8727. Holiday 1965 {published data only} Holiday AR, Joffe JR. A controlled evaluation of protriptyline conpared to a placebo and to methylphenidate hydrochloride. The Journal of New Drugs 1965;5:2578. Hornig-Rohan 2002 {published data only} Hornig-Rohan M, Amsterdam JD. Venlafaxine versus stimulant therapy in patients with ADD and depression. Progress in Neuropsychophramacology 2002;26:5859. Hunter 1967 {published data only} Hunter JW, Owen S, Eksi A, Scott JF. A controlled crossover study of trimipramine and amylobarbitone. British Journal of Psychiatry 1967;113:66770. Insel 1983a {published data only} Insel TR, Hamilton JA. Guttmacher LB. Murphy DL. D-amphetamine in obsessive-compulsive disorder. Psychopharmacology 1983;80:2315. Insel TR, Mueller ES, Gillin JC, Siever LJ, Murphy DL. Tricyclic response in obsessive compulsive disorder. Progress in Neuro-Psychopharmacology and Biological Psychiatry 1985; 9:2531. Insel TR, Murphy DL, Cohen RM, Alterman I, Kilts C, Linnoila M. Obsessive-compulsive disorder. A double-blind trial of clomipramine and clorgyline. Archives of General Psychiatry 1983;40:60512. Jefferson JW, Greist JH, Perse TL, Rosenfeld R. Fluvoxamine associated mania/hypomania in patients with obsessive compulsive disorder. Journal of Clinical Psychopharmacology 1991;11:3912. Zahn TP, Insel TR, Murphy DL. Psychophysiological changes during pharmacological treatment of patients with
15
obsessive compulsive disorder. British Journal of Psychiatry 1984;145:3944. Jacobson 1958 {published data only} Jacobson A. The use of ritalin in psychotherapy of depressions of the aged. Psychiatric Quarterly 1958;32: 47483. Joffe 1991 {published data only} Joffe RT, Swinson RP, Levitt AJ. Acute psychostimulant challenge in primary obsessive-compulsive disorder. Journal of Clinical Psychopharmacology 1991;11:23741. Kaplitz 1975 {published data only} Kapltz SE. Withdrawn, apathetic geriatric patients responsive to methylphenidate. Journal of the American Geriatrics Society 1975;23:2716. Kerenyi 1960 {published data only} Kerenyi AB, Koranyi EK, Sarwer-Foner GJ. Depressive states and drugs-III. Use of methylphenidate in open psychiatric settings in ofce practice. Canadian Medical Association Journal 1960;83:124954. Klein 1991 {published data only} Klein E, Zohar J, Geraci MF, Murphy DL, Uhde TW. Anxiogenic effects of m-CPP in patients with panic disorder: comparison to caffeines anxiogenic effects. Biological Psychiatry 1991;30:97384. Lader 1965 {published data only} Lader MH, Bond AJ, James DC. Clinical comparison of anxiolytic drug therapy. Psychological Medicine 1974;4: 3817. Lader 1974 {published data only} Bond AJ, James DC, Lader MH. Sedative effects on physiological and psychological measures in anxious patients. Psychological Medicine 1974;4:37480. Landman 1958 {published data only} Landman ME, Preisig R, Perlman M. A practical mood stimulant. The Journal of the Medical Society of New Jersey 1958;55:558. Leonard 2004 {published data only} Leonard BE, McCartan D, White J, King DJ. Methylphenidate: a review of its neuropharmacological, neuropsychological and adverse clinical effects. Human Psychopharmacology 2004;19:15180. Little 1989 {published data only} Little KY. Detecting acute drug effects. Biological Psychiatry 1989;25:6457. Little 1993 {published data only} Little KY. D-Amphetamine versus methylphenidate effects in depressed inpatients. Journal of Clinical Psychiatry 1993; 54:5257. London 1966 {published data only} London AM, Schreiber ED. A controlled study of effects of group discussions and an anorexiant in outpatient treatment of obesity. Internal Medicine 65;1:8092. Lovett Doust 1959 {published data only} Lovett Doust JW, Lewis DJ, Miller A, Sprott D, Wright RLD. Controlled asessment of antidepressant drugs
including tofranil. Canadian Psychiatric Association Journal 1959;4:S191S194. Masand 1991 {published data only} Masand P, Pickett P, Murray GB. Psychostimulants for secondary depression in medical illness. Psychosomatics 1991;32:2038. McDowell 2000 {published data only} McDowell DM, Levin FR, Seracini AM, Nunes EV. Venlafaxine treatment of cocaine abusers with depressive disorders. American Journal of Drug and Alcohol Abuse 2000; 26:2531. Meston 2004 {published data only} Meston CM. A randomized, placebo-controlled, crossover study of ephedrine for SSRI-induced female sexual dysfunction. Journal of Sex and Marital Therapy 2004;30: 5768. Natale 1979 {published data only} Natale M, Dahlberg CC. The effect of psychotomimetics on therapist patient matching of speech rhythms. Journal of Communication Disorders 1979;12:4552. Naylor 1986 {published data only} Naylor GJ, Martin B, Hopwood SE, Watson Y. A two-year double-blind crossover trial of the prophylactic effect of methylene blue in manic-depressive psychosis. Biological Psychiatry 1986;21:91520. Naylor 1987 {published data only} Naylor GJ, Smith AH, Connelly P. A controlled trial of methylene blue in severe depressive illness. Biological Psychiatry 1987;22:6579. Niederhofer 2002 {published data only} Niederhofer H. Therapy resistant major depression: symptom improvement psychopharmacological combination of antidepressants and methyphenidate. Wiener Medizininische Wochenschrift 2002;152:57880. Padala 2005 {published data only} Padala PR, Petty F, Bhatia SC. Methylphenidate may treat apathy independent depression. Annals of Pharmacotherpy 2005;39:19479. Pataki 1993 {published data only} Pataki CS, Carlson GA, Kelly KL, Rapport MD, Biancaniello TM. Side effects of Methylphenidate and Desipramine alone and in combination in children.. Journal of the American Academy of Child and Adolescent Psychiatry 1993;32:106572. Post 1974 {published data only} Post RM, Gillin JC, Wyatt RJ, Goodwin FK. The effect of orally adminstered cocaine on sleep of depressed patients. Psychopharmacologia 1974;37:5966. Quadri 1980 {published data only} Quadri AA, Shalini K, Channabasavanna SM. DAmphetamine as a predictor for response to imipramine and amitriptyline. Indian Journal of Psychiatry 1980;22:1824. Randall 2005 {published data only} Randall DC, Cafferty FH, Shneerson JM, Smith IE, Llewelyn MB, File SE. Chronic treatment with modanil
16
may not be benecial in patients with chronic fatigue syndrome. Journal of Psychopharmacology 2005;19:64760. Rapport 1993 {published data only} Rapport MD, Carlson GA, Kelly KL, Pataki C. Methylphenidate and desipramine in hospitalised children. Journal of American Academic Child and Adolescent Psychiatry 1993;32:33341. Rees 1961 {published data only} Rees L, Davies B. A controlled trial of phenelzine in the treatment of severe depressive illness. Journal of Mental Science 1961;107:560. Riddle 1995 {published data only} Riddle MA, Lynch KA, Scahill L, de Vries A, Cohen DJ, Leckman JF. Methylphenidate discontinuation and reinintiation during long-term treatment of children with Tourettes Disorder and attention decit disorder: a pilot study. Journal of Child and Adolescent Psychopharmacology 1995;5:20514. Rosenquist 1994 {published data only} Rosenquist PB, McCall WV, Farah A, Reboussin DM. Effects of caffeine pretreatment on measures of seizure impact.. Convulsive Therapy 1994;10:1815. Roy-Byrne 1983 {published data only} Roy-Byrne PP, Gwirtsman HE, Baxter L, Gerner RH. Endocrine correlates of mood response to methylphenidate in depression. Journal of Clinical Psychopharmacology 1983; 3:2668. Schmitz 2001 {published data only} Schmitz JM, Averill P, Stotts AL, Moeller FG, Rhoades HM, Grabowski J. Fluoxetine treatment of cocaine-dependent patients with major depressive disorder. Drug and Alcohol Dependence 2001;63:20714. Sramek 1995 {published data only} Sramek JJ, Jasinsky O, Chang S, Shu V, Kashkin K, Kennedy S, et al.Pilot efcacy trial of ABT-200 in patients with major depressive disorder. Depression 1995;2:3158. Tremblay 2002 {published data only} Tremblay LK, Naranjo CA, Cardenas L, Herrmann N, Busto UE. Probing brain reward system function in major depressive disorder: altered response to dextroamphetamine.. Archives of General Psychiatry 2002; 59:40916. Tremblay 2002b {published data only} Cardenas L, Tremblay LK, Naranjo CA, Herrmann N, Zack M, Busto UE. Brain reward system activity in major depression and comorbid nicotine dependence. Journal of Pharmacology and Experimental Therapeutics 2002.;302: 126571. Van Kammen 1975a {published data only} Van Kammen DP, Murphy DL. Attenuation of the euphoriant and activating effects of d- and l-Amephetamine by lithium carbonate treatment. Psychopharmacologia 1975; 44:21524.
Van Kammen 1975b {published data only} Gillin JC, Van Kammen DP, Graves J, Murphy D. Differential effects of D and L amphetamine on the sleep of depressed patients. Life Sciences 1975;17:123340. Van Kammen 1978 {published data only} Van Kammen DP, Murphy DL. Prediction of Imipramine antidepressant response by a one-day d-amphetamine trial. American Journal of Psychiatry 1978;135:117984. Van Praag 1973 {published data only} Van Praag HM, Korf J. 4-Chlorampetamines. Chance and Trend in the development of new antidepressants. Journal of Clinical Pharmacology and New drugs 1973;13:314. Wheatley 1969c {published data only} Wheatley D. Amphetamines in general practice: their use in depression and anxiety. Seminars in Psychiatry 1969;1: 16373. Woods 1986 {published data only} Woods SW, Tesar GE, Murray GB, Cassem NH. Psychostimulant treatment of depressive disorders secondary to medical illness. Journal of Clinical Psychiatry 1986;47: 125. Ziedonis 1991a {published data only} Ziedonis DM, Kosten TR. Depression as a prognostic factor for pharmacolgical treatment of cocaine dependence. Psychopharmacology Bulletin 1991;27:33743. Ziedonis 1991b {published data only} Ziedonis DM, Kosten TR. Pharmacotherapy improves treatment outcome in depressed cocaine addicts. Journal of Psychoactive Drugs 1991;23:41725.
Additional references
APA 2005 American Psychiatric Assocaition. Diagnostic and statisticsl manual of mental disorders (DSM-5). Washington, DC: Americian Psychiatric Association, 2005. Bebbington 2000 Bebbington PE, Brugha TS, Meltzer H, Jenkins R, Ceresa C, Farell M, et al.Neurotic disorders and the recipt of psychiatric treatment.. Psychological Medicine 2000;30: 136976. BNF 2007 British Medical Association and Royal Pharmaecutical Society of Great Britian.. British National Formulary 53rd Edition. London: British Medical Assocaition and Royal Pharmaecutical Society od Great Britian, 2007. Bosc 1997 Bosc M, Dubini A, Polin V. Development and validation of a social functioning scale, the Social Adaptation Selfevaluation Scale. European Neuropsychopharmacology 1997; 7:S57 S79. Cassano 2002 Cassano P, Fava M. Depression and public health. An overview.. Journal of Psychosomatic Research 2002;53: 84957.
17
Chiarello 1987 Chiarello RJ, Cole JO. The use of psychostimulants in general psychiatry. Archives of General Psychiatry 1987;44: 28695. Delgado 2000 Delgado PL. Depression: the case for a monoamine deciency. Journal of Clinical Psychiatry 2000;61 Suppl 6: 711. Demyttenaere 2004 Demyttenaere K, Bruffaerts R, Posada-Villa J, Gasquet I, Kovess V, Lepine JP, et al.Prevalence, severity and unmet need for treatment of mental disorders in the World Health Organisation World Mental Health Surveys.. JAMA 2004; 291:258190. Hamilton 1960 Hamilton M. A rating scale for depression. Journal of Neurology, Neurosurgy and Psychiatry 1960;23:5662. Higgins 2006 Higgins JPT, Greens S, editors. Cochrane Handbook for Systematic Reviews of Interventions 4.2.6 (updated September 2006). www.cochrane.org/resoures/handbook/ hbook.htm (accessed September 30th 2006). Wiley, 2006. Juni 2001 Juni P, Altman DG, Egger M. Systematic reviews in health care: Assessing the quality of controlled clincal trials. BMJ 2001;323:426. Lin 1999 Lin E, Parikh SV. Sociodemographic, clinical, and attitudinal characteristics of the untreated depressed in Ontario. Journal of Affective Disorders 1999;53:15362. Linet 1989 Linet LS. Treatment of a refractory depression with a coombination of uoxetine and d-amphetamine. American Journal of Psychiatry 1989;146:8034. Maddox 1994 Maddox JC, Levi M, Thompson C. The compliance with antidepressants in general practice.. Journal of Psychopharmacology 1994;8:4853. Masand 1998 Masand PS, Anand VS, Tanquary JF. Psychostimulant augmentation of second generation anti-depressants: A case series. Depression and Anxiety 1998;7(2):8991. Moncrieff 2005 Moncrieff J, Kirsch I. Efcacy of antidepressants in adults. BMJ 2005;331:1557. Montgomery 1979 Montgomery SA, Asberg M. A new depression scale designed to be sensitive to change. British Journal of Psychiatry 1979;134:3829.
Murray 1997 Murray CJ, Lopez AD. Alternative projections of mortality and disability by cause 1990-2020: Golbal Burden of Disease Study. Lancet 1997;349:149804. Nguyen 2005 Nguyen B, Tampi RR. Psychostimulants in the treatment of depression in the older patient. Clinical Geriatrics 2005;13: 3945. NICE 2004 National Institute for Clinical Excellence. Depression: management of depression in primary and secondary care. Clinical practice guideline No 23. London: National Institute for Clinical Excellence, 2004. Orr 2007 Orr K, Taylor D. Psychostimulants in the treatment of depression. CNS Drugs 2007;21:23957. Satel 1989 Satel SL, Nelson JC. Stimulants in the treatment of depression: a critical overview. Journal of Clinical Psychiatry 1989;50:24912. Singleton 2001 Singleton N, Bumpstead R, OBrien M, Lee A, Meltzer H. Psychiatric morbidity among adults living in private households, 2000.. London: The Stationery Ofce, 2001. Spitzer 1995 Spitzer RL, Kroenke K, Linzer M, Hahn SR, Williams JB, deGruy FV 3rd, et al.Health-related quality of life in primary care patients with mental disorders.. JAMA 1995; 274:15117. Thomas 2003 Thomas CM, Morris S. Cost of depression among adults in England in 2000. British Journal of Psychiatry 2003;183: 5149. Wang 2005 Wang PS, Lane M, Olfson M, Pincus HA, Wells KB, Kessler RC. Twelve-month use of mental health services in the United States.. Archives of General Psychiatry 2005;62: 62940. Ware 1992 Ware JE Jr, Sherbourne CD. The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Medical Care 1992;30:47383. Young 2001 Young AS, Klap R, Sherbourne CD, Wells KB. The quality of care for depressive and anxiety disorders in the United Sates.. Archives of General Psychiatry 2001;58:5561. Indicates the major publication for the study
18
CHARACTERISTICS OF STUDIES
Characteristics of included studies [ordered by study ID]

Cantello 1989 Methods Participants Double blind randomised controlled cross-over trial Four armed trial. The two arms are relevant to this review. Arm 1- Patients with major depression (DSM-III) N=14 male (n=8) and female patients. Arm 2- Patients with major depression (DSM-III) and Parkinsonian N=13 male (n=7) and female patients Mean age in those depressed was 59 years, in depressed patients with Parkinsonians disease it was 57 years Italy Intravenous one dose each of: Drug 1 = methylamphetamine 0.4 mg/kg. Drug 2 = placebo Duration of treatment: one dose of each drug separated by 3 days Self- rated depressed effect plus dysphoria, Mood- observers global judgement score
Interventions
Outcomes Notes Risk of bias Item Allocation concealment? Cookson 1986 Methods Participants
Authors judgement Unclear
Description B - Unclear
Double blind randomised controlled cross-over trial N=21 male (n=5) and female patients with depressive disorder (RDC) Age range 23-74 UK Intravenous one dose each of: Drug 1 = methylamphetamine 15 mg . Drug 2 = placebo Duration of treatment: 2 days unclear if treatments separated (for washout) Self report using visual analogue scale (VAS) for depression, anxiety and hunger at 1,2,4 and 6 hours after. Author rating of the overall severity of depression using VAS
Interventions
Outcomes
Notes Risk of bias
19
Cookson 1986
(Continued)
Item Allocation concealment? DeBattista 2003 Methods Participants
Double blind randomised controlled parallel multi-centred study N= 136 male (n=41) and female patients with major depression (DSM-IV) and partial response to antidepressant therapy. Mean age 45 years. USA Oral Group 1 = modanil once daily 100-400 mg Group 2 = placebo In addition both groups received as part of their regular treatment an unspecied anti-depressant Duration of treatment: 6 weeks Depression: Hamilton Depression Scale (HAM-D) 21. Clinical Global Impression of severity (CGI-S) and Clinical Global Impression of Change (CGI-C). Fatigue and sleepiness using the Fatigue Severity Scale and the Epworth Sleepiness Scale
Interventions
Outcomes
Notes Risk of bias Item Allocation concealment? Elizur 1979 Methods Participants Double blind randomised controlled parallel trial N=20 male (n=10) and female psychiatric outpatients suffering from major depression (diagnostic measure not stated) Mean age 36.5 years in pemoline group and 33.7 in placebo group (range 30-60) Israel Oral Group 1 = pemoline 50mg daily Group 2 = placebo Duration of treatment: 3 weeks Modied HAM-D (using 8 target symptoms) and CGI Authors judgement Unclear Description B - Unclear
Interventions
Outcomes Notes
20
Elizur 1979
(Continued)
Risk of bias Item Allocation concealment? Fava 2005 Methods Participants Double blind randomised controlled parallel multi-centred study N=314 male (n = 94) and female patients diagnosed with MDD (DSM-IV) , with symptoms of fatigue and excessive sleepiness and were partial responders to antidepressant therapy Mean age 42 years (18-65 years) USA Oral Group 1 = modanil 100-200mg daily Group 2 = placebo Patients continued to recieve as part of their regular treatment a selective serotonin reuptake inhibitor (SSRIs) antidepressant Duration of treatment: 8 weeks Depression, sleepinesss, fatigue, and depression. Using the HAM-D 31 and 17, Montgomery-Asberg depression rating scale (MADRS), CGI-S, CGI-I, Brief Fatigue Inventory, and Epworth Sleepiness Scale Authors judgement Unclear Description B - Unclear
Interventions
Outcomes
Notes Risk of bias Item Allocation concealment? Fernandez 1995 Methods Participants Double blind randomised controlled parallel trial N= 20 male patients with major depressive symptoms (DSM-III-R) and HIV Mean age in psychostimulant group: 35.5 years (SD 10.48), in anti-depressant group 2: 34 years (SD 7. 91) USA Oral Group 1 = methylphenidate 5-30mg daily Group 2 = desipamine 25mg-150mg daily Duration of treatment : 6 weeks
21
Interventions
Fernandez 1995
(Continued)
Outcomes Notes Risk of bias Item Allocation concealment? Hare 1962 Methods Participants
HAM-D 24, Brief Symtpom Inventory Depression scale (BSI), State-Trait Anxiety Inventory (STAI)
Double blind randomised controlled cross-over trial N= 46 male (n=12) and female patients diagnosed with primary depressive illness (diagnostic measure not stated) Median age of completers 42 years UK Oral Twice daily per drug. Drug 1 = dexamphetamine 5mg Drug 2 = phenezine 30mg Drug 3 placebo Duration of treatment: 2 weeks on weekdays only Assessor rating of depression, retardation, agitation, anxiety, hypochondria, anorexia and insomnia
Interventions
Outcomes Notes Risk of bias Item Allocation concealment? Hare 1964 Methods Participants
Double blind randomised controlled cross-over trial N= 106 male and female (does not provide baseline proportion) in/day patients diagnosed with primary depressive illness. 32 male and 46 females completed study Median age of completers 45 years. UK Oral Drug 1 = dexamphetamine 5 mg and amylobarbitone 50 mg Drug 2 = imipramine 75mg increased to 150 mg.
22
Interventions
Hare 1964
(Continued)
Duration of treatment: Over 6 weeks, 3 weeks per drug. Outcomes Notes Risk of bias Item Allocation concealment? Kits 1970 Methods Participants Double blind randomised controlled parallel trial N= 50 male (n=25) and female inpatients, hospitalised for depressive symptoms Mean age 44 years, range 19-69. The Netherlands Oral Group 1 = p-chloro-n-methylamphetamine 90 mg daily Group 2 = placebo Duration of treatment: 4 weeks Vital syndrome interview- severity of depression, mood, sleep, appetite. Beck Depression Inventory and Zung self-rating scale Authors judgement Unclear Description B - Unclear Assessor rating of depression, anxiety, insomnia (subject rating) and overall clinical state
Interventions
Outcomes
Notes Risk of bias Item Allocation concealment? Lee 2005 Methods Participants Double blind randomised controlled parallel trial N=30 male (n=24) and female patients with traumatic brain injury with major depression (DSM-IV) Mean age of completers in methylphenidate group 35.3 (SD 8.0), in setraline group 33.6 (SD 12.3) and in placebo group 35.5 (SD 7.2) years Korea Oral Group 1 = methylphenidate to 20mg daily Group 2 = placebo
23
Description D - Not used
Interventions
Lee 2005
(Continued)
Group 3 = sertraline to 100mg daily Duration of treatment 1 week Outcomes Notes Risk of bias Item Allocation concealment? Mattes 1985 Methods Participants Double blind randomised controlled cross-over trial N=20 male (n=10) and female outpatients mildly depressed (HAM-D, SCL-90) Mean age 30.7 (range 22-45) USA Oral Drug 1 = methylphenidate 10 mg up to 60mg daily Drug 2 = placebo Duration of treatment unclear either 3 weeks overall or 3 weeks per treatment POMS, Symptom Distress Checklist-90, California Psychological Inventory, SAD-C (global assessment scale and scales for depressive syndrome) Authors judgement Unclear Description B - Unclear HAM-D-17, Beck Depression Inventory
Interventions
Outcomes
Notes Risk of bias Item Allocation concealment? Overall 1962 Methods Participants Double blind randomised controlled parallel multi-centred trial N=204 male (n=194) and female in-patients with a main psychiatric diagnosis of either (1) neurotic depresive reaction, (2) psychotic depressive reaction, (3) manic-depressive reaction , depressed type, (4) involutional psychotic reaction, depressive type, or (5) schizophrenic reaction, schizo-affective type, depressed Age not provided. USA Authors judgement Unclear Description B - Unclear
24
Overall 1962
(Continued)
Interventions
Oral Group 1 = imipramine 37.5mg TO 300 mg Group 2 = isocarboxazide 5mg to 40 mg Group 3 = dextroamphetamine 5 mg tp 40 mg and amobarbital 32mg to 256 mg daily Group 4 = placebo Treatment duration: 12 weeks IMPS-31
Outcomes Notes Risk of bias Item Allocation concealment? Patkar 2006 Methods Participants
Double blind randomised controlled parallel trial N=60 male (states about 37% of completers were male) and female outpatients with treatment resistant major depressive disorder (DSM-IV) Mean age of copmpleters was 48.5 years. USA Oral Group 1 = methylphenidate 18-54 mg daily Group 2 = placebo Patients remain on preexisiting antidepressant Duration of treatment: 4 weeks HAM-D 21, CGI-S, CGI-I, Beck Depression Inventory (BDI-11)
Interventions
Outcomes Notes Risk of bias Item Allocation concealment?
25
Postolache 1999 Methods Participants Double blind randomised controlled parallel trial N= 9 male (n = 2) and female out-patients with major depressive disorder (DSM-IV)> Age range 24-66 years USA Oral Group 1 = methylphenidate increasing up to 10 mg daily Group 2 = placebo All patients remained on pre-existing antidepressant of sertraline 200mg daily Duration of treatment: 9 weeks HAM-D 21
Interventions
Outcomes Notes Risk of bias Item Allocation concealment? Reus 1979 Methods Participants
Double blind randomised controlled cross-over trial N= 9 endogenously depressed (8 unipolar and 1 bipolar) (RDC) and had concomitant fatgue or apathy. Does not state if male or female Age range 21-40 years. USA Intravenous Group 1 = dextroamphetamine 20mg Group 2 = placebo Duration of treatment: one dose separated by 2 days Brief Psychiatric Rating Scale Modied Bunney-Hamburg Scale Self-rated depression, activated euphoria, anxiety
Interventions
Outcomes
Notes Risk of bias Item Allocation concealment? Authors judgement Unclear Description B - Unclear
26
Rickels 1970 Methods Participants Double blind randomised controlled parallel trial N = 120 male (n=23) and female outpatient , private psychiatric practice patients or general practice patients that were mild to moderately depressed (GP or psychiatrist diagnosis) with symptoms of fatigue, apathy or anorexia Mean age of those recruited from clinc 41 years, from general practice 56 years and from psychiatric private practice 38 years USA Oral Group 1 = pemoline 25 mg Group 2 = methylphenidate 5 mg 3 times daily Group 3 = placebo Duration of treatment: 4 weeks Zung Depression Scale Physican rated fatigue, anorexia, anxiety, depression. sleep disturbances, appetite disturbances
Interventions
Outcomes
Notes Risk of bias Item Allocation concealment? Rickels 1972 Methods Participants Double blind randomised controlled parallel trial N= 101 male (n=33) and female outpatients that were mildly depressed (GP diagnosed) Mean age 32.7yrs USA Oral Group 1 = methylphenidate 30mg daily (could be decreased or increased to 60mg daily) Group 2 = placebo Duration of treatment: 4 weeks Zung Depression Scale Physician depression scale Authors judgement Unclear Description B - Unclear
Interventions
Outcomes
27
Robin 1958 Methods Participants Double blind randomised controlled parallel trial N= 45 male (n=14) and female patients with mild or moderate depression (psychiatrist diagnosed). All participants had at least 4 other symptoms of sleep disturbance, loss of appetite and weight loss, impairment of interests, lack of condence, reduced alertness and retardation or agitation Mean age in intervention group 37.5 years, in placebo 39.5 years UK Oral Group 1 = methylphenidate 10-20 mg Group 2 = placebo Duration of treatment: 4 weeks Psychiatrist rating, Patient rated improvement, depressive scale of MMPI
Interventions
Outcomes Notes Risk of bias Item Allocation concealment? Silberman 1981 Methods Participants
Double blind randomised cross-over trial N = 20 male (n =10 of completers) and female patients with major depressive episode (RDC) Mean age of completers 33.4 years USA Intravenous One dose of Drug 1 = dextroamphetamine 20mg Drug 2 = placebo infusion Duration of treatment: one dose each cross over separated by at least 2 days Physician BPRS
Interventions
28
Vaishnavi 2006 Methods Participants Double blind randomised controlled parallel trial N = 50 male (n =7) and female patients with major atypical depressive disorder (DSM-IV). (atypical described as a subgroup of depressed patients who have reactive mood with features of hypersomnia, hyperphagia, anergia and regection sensitivity and where convention antidepressant therpay is of limited value Mean age 39.48 +/- 9.44 USA Oral Group 1 = modanil 200mg/d to 400 mg/d Group 2 = placebo Duration of treatment 12 weeks HAM-D 29, SCL-90 anxiety, depression and over eating subscales, CGI-S, FSS, BFI, ADDS
Interventions
Outcomes Notes Risk of bias Item Allocation concealment? Wagner 2000 Methods Participants
Double blind randomised controlled parallel trial N =23 male patients with depressive disorder (DSM-IV) AND hiv Mean age 41 years (SD 8) USA Oral Group 1 = dextroamphetamine increased up to maximum dose of 40mg per day. Group 2 = placebo Duration of treatment 2 weeks HAM-D-21, Brief Symptom Inventory (BSI) Clinical Global Impressions, 10 point VAS depression, 7 items CFS, 10 point VAS energy
Interventions
Outcomes
29
Wallace 1995 Methods Participants Double blind randomised controlled cross-over trial N = 16 male (n = 5) and female patients with major depression (DSM-III) Mean age 72.3 years (SD 9.8) USA Oral Group 1 = methylphenidate 10mg-20mg DAILY Group 2 = placebo Duration of treatment: 8 days HAM-D(does not state which version)
Interventions
Outcomes Notes Risk of bias Item Allocation concealment? Wheatley 1969 a Methods Participants
Double blind randomised controlled parellel trial N = 70 male (n=9) and female patients with acute depression. How diagnosed not stated No patients were aged less than 16 or greater than 70 years UK Oral Group 1 = dexamphetamine 10mg- 30mg daily Group 2 = placebo Duration of treatment 3 weeks Clinical assessments on symptoms improvement
Interventions
30
Wheatley 1969 b Methods Participants Double blind randomised controlled cross-over trial N = 40 male (n = 7) and female patients with chronic depresssion. How depression diagnosed unclear No patients aged less than 16 years UK Oral Group 1 = dextroamphetamine 10mg - 30 mg daily Group 2 = placebo Duration of treatment 8 weeks Clinical assessments on on symptom improvement
Interventions
Characteristics of excluded studies [ordered by study ID]
Study Ballinger 1974 Blashki 1971 Brooks 2003 Buchsbaum 1977 Buckman 1973 Camacho 2006 Capstick 1965 Carlson 1995 Carlson 2004 Carroll 1995 Christensen 1990
Reason for exclusion Not psychostimulants Not psychostimulants Not psychostimulants No depression outcome Not psychostimulants Case report Not psychostimulants Case report Case series Not psychostimulants Not psychostimulants
31
(Continued)
Ciraulo 2005 Clauson 2004 Coffey 1990 Cornelius 1994 Daley 1998 Darvill 1959 de Wit 1987 Duval 1996 Fava 2006 Fawcett 1971 Fawcett 1972 Fawcett 1987 Giannini 1986 Grade 1998 Harris 1960 Hicks 1994 Hinton 1959 Holiday 1965 Hornig-Rohan 2002 Hunter 1967 Insel 1983a Jacobson 1958 Joffe 1991 Kaplitz 1975
Not psychostimulants Case series No depression outcome Not psychostimulants Not psychostimulants No depression outcome Not a RCT evaluation Not psychostimulants Review article Not a RCT evaluation Not a RCT evaluation Not psychostimulants Not psychostimulants Patients were not diagnosed with depression Not psychostimulants No depression outcome Not psychostimulants Not a RCT evaluation Not a RCT evaluation Not psychostimulants Not psychostimulants Not a RCT evaluation No depression outcome No depression outcome
32
(Continued)
Kerenyi 1960 Klein 1991 Lader 1965 Lader 1974 Landman 1958 Leonard 2004 Little 1989 Little 1993 London 1966 Lovett Doust 1959 Masand 1991 McDowell 2000 Meston 2004 Natale 1979 Naylor 1986 Naylor 1987 Niederhofer 2002 Padala 2005 Pataki 1993 Post 1974 Quadri 1980 Randall 2005 Rapport 1993 Rees 1961
Not a RCT evaluation Patients were not diagnosed with depression at baseline Not psychostimulants Not psychostimulants Not a RCT evaluation Review article A validation study Not a RCT evaluation Patients were not diagnosed with depression Not a RCT evaluation Not a RCT evaluation Not psychostimulants No depression outcome No depression outcome Not psychostimulants Not psychostimulants Case report Case report Evalaution in children No depression outcome Not a RCT evaluation Patients were not diagnosed with depression at baseline Evaluation in children Not psychostimulants
33
(Continued)
Riddle 1995 Rosenquist 1994 Roy-Byrne 1983 Schmitz 2001 Sramek 1995 Tremblay 2002 Tremblay 2002b Van Kammen 1975a Van Kammen 1975b Van Kammen 1978 Van Praag 1973 Wheatley 1969c Woods 1986 Ziedonis 1991a Ziedonis 1991b
Evaluation in children No depression outcome Not a RCT evaluation Not psychostimulants Not psychostimulants No depression outcome Cardenas 2002: Not specically looking at the effect of d-amphetamines on depression Not a RCT evaluation No depression outcome Not a RCT evaluation Not a RCT evaluation Outcome does not measure effectiveness of PS Not a RCT evaluation Not psychostimulants Not psychostimulants
34
DATA AND ANALYSES
Comparison 1. Psychostimulants versus placebo
Outcome or subgroup title 1 Reduction in depression symptoms (short-term) 1.1 Psychostimulants 1.2 Modanil 2 Reduction in depression symptoms (medium term) 2.1 Psychostimulants 2.2 Modanil 3 Clinical response (short term) any measure/denition 3.1 Psychostimulants 3.2 Modanil 4 Improvement in cognitive functioning (short term) 4.1 Psychostimulants 4.2 Modanil 5 Reduction in fatigue (short term) 5.1 Psychostimulants 5.2 Modanil 6 Reduction in fatigue (medium term) 6.1 Psychostimulants 6.2 Modanil 7 Reduction in hypersomnia (medium term): Epworth Sleepiness Scale 7.1 Psychostimulants 7.2 Modanil 8 Improvement in health-related quality of life (short term) 8.1 Psychostimulants 8.2 Modanil 9 Acceptability: dropouts due to side effects (short term) 9.1 Psychostimulants 9.2 Modanil 10 Tolerability: proportion of participants experiencing side effects (short term) 10.1 Psychostimulants 10.2 Modanil
No. of studies 3 3 0 1 0 1 3 3 0 2 2 0 2 2 0 1 0 1 1
No. of participants
Statistical method Std. Mean Difference (IV, Fixed, 95% CI)
Effect size Subtotals only -0.87 [-1.40, -0.33] Not estimable Subtotals only Not estimable -2.7 [-7.76, 2.36] Subtotals only 1.01 [0.48, 2.09] Not estimable Subtotals only 0.42 [-0.20, 1.04] Not estimable Subtotals only -1.80 [-2.54, -1.06] Not estimable Subtotals only Not estimable 0.24 [-1.44, 1.92] Subtotals only
62 0
Std. Mean Difference (IV, Fixed, 95% CI) Std. Mean Difference (IV, Fixed, 95% CI) Mean Difference (IV, Fixed, 95% CI) Mean Difference (IV, Fixed, 95% CI) Mean Difference (IV, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Std. Mean Difference (IV, Fixed, 95% CI) Std. Mean Difference (IV, Fixed, 95% CI) Std. Mean Difference (IV, Fixed, 95% CI) Std. Mean Difference (IV, Fixed, 95% CI) Std. Mean Difference (IV, Fixed, 95% CI) Std. Mean Difference (IV, Fixed, 95% CI) Mean Difference (IV, Fixed, 95% CI) Mean Difference (IV, Fixed, 95% CI) Mean Difference (IV, Fixed, 95% CI) Mean Difference (IV, Fixed, 95% CI)
0 50
130 0
42 0 42 0
0 41
0 1 2 2 0 2 2 0 2
0 41
Mean Difference (IV, Fixed, 95% CI) Mean Difference (IV, Fixed, 95% CI) Std. Mean Difference (IV, Fixed, 95% CI) Std. Mean Difference (IV, Fixed, 95% CI) Std. Mean Difference (IV, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Random, 95% CI)
Not estimable -0.24 [-3.06, 2.58] Subtotals only 0.57 [-0.05, 1.20] Not estimable Subtotals only 1.63 [0.26, 10.25] Not estimable Subtotals only
42 0
68 0
2 0
185 0
Odds Ratio (M-H, Random, 95% CI) Odds Ratio (M-H, Random, 95% CI)
1.78 [0.57, 5.55] Not estimable

35
11 Tolerability: proportion of participants experiencing side effects (medium term) 11.1 Psychostimulants 11.2 Modanil 12 Tolerability: dizziness (short term) 12.1 Psychostimulants 12.2 Modanil 13 Tolerability: drowsiness (short term) 13.1 Psychostimulants 13.2 Modanil 14 Tolerability: headache (short term) 14.1 Psychostimulants 14.2 Modanil 15 Tolerability: insomnia (short term) 15.1 Psychostimulants 15.2 Modanil 16 Tolerability: nausea (short term) 16.1 Psychostimulants 16.2 Modanil 17 Tolerability: tremor (short term) 17.1 Psychostimulants 17.2 Modanil
Odds Ratio (M-H, Random, 95% CI)
Subtotals only
2 0 2 2 0 2 2 0 2 2 0 1 1 0 2 2 0 2 2 0
90 0
Odds Ratio (M-H, Random, 95% CI) Odds Ratio (M-H, Random, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI)
7.22 [2.21, 23.57] Not estimable Subtotals only 1.31 [0.40, 4.24] Not estimable Subtotals only 0.41 [0.12, 1.37] Not estimable Subtotals only 1.06 [0.34, 3.35] Not estimable Subtotals only 1.89 [0.66, 5.38] Not estimable Subtotals only 1.52 [0.31, 7.57] Not estimable Subtotals only 2.83 [0.51, 15.67] Not estimable
190 0
190 0
115 0
95 0
190 0
115 0
Comparison 2. Psychostimulants versus antidepressants
Outcome or subgroup title 1 Reduction in depression symptoms (short term) 1.1 Psychostimulants 1.2 Modanil 2 Tolerability: number of participants experiencing side effects (medium term) 2.1 Psychostimulants 2.2 Modanil
No. of studies 1 1 0 1
No. of participants
Statistical method Mean Difference (IV, Fixed, 95% CI)
Effect size Subtotals only -4.30 [-8.79, 0.19] Not estimable Subtotals only
20 0
Mean Difference (IV, Fixed, 95% CI) Mean Difference (IV, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI)
1 0
15 0
Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI)
1.43 [0.10, 20.44] Not estimable
36
Comparison 3. Psychostimulants versus placebo as adjunct to antidepressant treatment
Outcome or subgroup title 1 Reduction in depression symptoms (short term) 1.1 Psychostimulants 1.2 Modanil 2 Reduction in depression symptoms (medium term) 2.1 Psychostimulants 2.2 Modanil 3 Clinical response (short term) 3.1 Psychostimulants 3.2 Modanil 4 Reduction in fatigue symptoms (short term) 4.1 Psychostimulants 4.2 Modanil 5 Reduction in fatigue symptoms (medium term) 5.1 Psychostimulants 5.2 Modanil 6 Reduction in hypersomnia (short term): Epworth Sleepiness Scale 6.1 Psychostimulants 6.2 Modanil 7 Reduction in hypersomnia (medium term): Epworth Sleepiness Scale 7.1 Psychostimulants 7.2 Modanil 8 Acceptability: dropouts due to side effects (short term) 8.1 Psychostimulants 8.2 Modanil 9 Acceptability: dropouts due to side effects (medium term) 9.1 Psychostimulants 9.2 Modanil 10 Acceptability: dropouts due to lack of efcacy (medium term) 10.1 Psychostimulants 10.2 Modanil 11 Tolerability: proportion of participants experiencing side effects (short term) 11.1 Psychostimulants 11.2 Modanil
No. of studies 3 1 2 2 0 2 1 1 0 2 0 2 2 0 2 2
No. of participants
Statistical method Mean Difference (IV, Fixed, 95% CI)
Effect size Subtotals only -1.60 [-5.96, 2.76] -0.09 [-1.07, 0.89] Subtotals only Not estimable -0.80 [-1.88, 0.28] Subtotals only 1.83 [0.57, 5.88] Not estimable Subtotals only Not estimable 0.04 [-0.79, 0.87] Subtotals only Not estimable -0.31 [-0.67, 0.05] Subtotals only
50 411
Mean Difference (IV, Fixed, 95% CI) Mean Difference (IV, Fixed, 95% CI) Mean Difference (IV, Fixed, 95% CI) Mean Difference (IV, Fixed, 95% CI) Mean Difference (IV, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Random, 95% CI) Mean Difference (IV, Fixed, 95% CI) Mean Difference (IV, Fixed, 95% CI) Mean Difference (IV, Fixed, 95% CI) Mean Difference (IV, Fixed, 95% CI)
0 443 50 0
0 437
0 443
0 2 2
0 437
Mean Difference (IV, Fixed, 95% CI) Mean Difference (IV, Fixed, 95% CI) Mean Difference (IV, Fixed, 95% CI)
Not estimable -0.79 [-1.63, 0.06] Subtotals only
0 2 1 1 0 3 1 2 2 0 2 1
0 443
Mean Difference (IV, Fixed, 95% CI) Mean Difference (IV, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI)
Not estimable -1.27 [-2.14, -0.39] Subtotals only 1.0 [0.13, 7.60] Not estimable Subtotals only 6.43 [0.23, 181.82] 1.31 [0.54, 3.18] Subtotals only Not estimable 0.97 [0.24, 3.96] Subtotals only
60 0
9 450
0 450
1 0
60 0
Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI)
1.32 [0.47, 3.72] Not estimable

37
12 Tolerability: headache (medium term) 12.1 Psychostimulants 12.2 Modanil 13 Tolerability: insomnia (short term) 13.1 Psychostimulants 13.2 Modanil 14 Tolerability: insomnia (medium term) 14.1 Psychostimulants 14.2 Modanil 15 Tolerability: nausea (short term) 15.1 Psychostimulants 15.2 Modanil 16 Tolerability: nausea (medium term) 16.1 Psychostimulants 16.2 Modanil 17 8Tolerability: tremor (short term) 17.1 Psychostimulants 17.2 Modanil 18 Tolerability: upper respiratory tract infection (medium term) 18.1 Psychostimulants 18.2 Modanil
2 0 2 1 1 0 2 0 2 1 1 0 2 0 2 1 1 0 2 0 2 0 447
Odds Ratio (M-H, Random, 95% CI) Odds Ratio (M-H, Random, 95% CI) Odds Ratio (M-H, Random, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Random, 95% CI) Odds Ratio (M-H, Random, 95% CI) Odds Ratio (M-H, Random, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI) Odds Ratio (M-H, Fixed, 95% CI)
Subtotals only Not estimable 1.21 [0.50, 2.94] Subtotals only 1.0 [0.19, 5.40] Not estimable Subtotals only Not estimable 1.24 [0.62, 2.50] Subtotals only 3.22 [0.32, 32.89] Not estimable Subtotals only Not estimable 1.78 [0.20, 15.95] Subtotals only 0.48 [0.04, 5.63] Not estimable Subtotals only Not estimable 0.73 [0.31, 1.71]
60 0
0 447
60 0
0 447
60 0
0 447
Comparison 4. Psychostimulants versus placebo: concomitant serious medical condition
Outcome or subgroup title 1 Reduction in depression symptoms (short term)
No. of studies 2
No. of participants
Statistical method Std. Mean Difference (IV, Fixed, 95% CI)
Effect size Subtotals only
38
Analysis 1.1. Comparison 1 Psychostimulants versus placebo, Outcome 1 Reduction in depression symptoms (short-term).
Review: Psychostimulants for depression
Comparison: 1 Psychostimulants versus placebo Outcome: 1 Reduction in depression symptoms (short-term)
Study or subgroup
Treatment N Mean(SD)
Control N Mean(SD)
Std. Mean Difference IV,Fixed,95% CI
Weight
1 Psychostimulants Elizur 1979 Lee 2005 Wagner 2000 10 10 11 2.5 (0.37) 15.7 (5.6) 7.5 (3.9) 10 10 11 3 (0.47) 22.3 (4.2) 9.4 (6.1) 30.8 % 29.4 % 39.8 % -1.13 [ -2.09, -0.17 ] -1.28 [ -2.26, -0.30 ] -0.36 [ -1.20, 0.49 ]
Subtotal (95% CI)
31
31
100.0 %
-0.87 [ -1.40, -0.33 ]
Heterogeneity: Chi2 = 2.37, df = 2 (P = 0.31); I2 =15% Test for overall effect: Z = 3.19 (P = 0.0014) 2 Modanil
Subtotal (95% CI)

Heterogeneity: not applicable Test for overall effect: not applicable
0.0 %
0.0 [ 0.0, 0.0 ]
Test for subgroup differences: Not applicable
-10
-5
10
Favours treatment
Favours control
39
Analysis 1.2. Comparison 1 Psychostimulants versus placebo, Outcome 2 Reduction in depression symptoms (medium term).
Comparison: 1 Psychostimulants versus placebo Outcome: 2 Reduction in depression symptoms (medium term)
Study or subgroup
Control N Mean(SD)
Mean Difference IV,Fixed,95% CI
Weight
1 Psychostimulants
Subtotal (95% CI)

Heterogeneity: not applicable Test for overall effect: not applicable 2 Modanil Vaishnavi 2006
0.0 %
0.0 [ 0.0, 0.0 ]
24
7.3 (7.1)
26
10 (10.9)
100.0 %
-2.70 [ -7.76, 2.36 ]
Subtotal (95% CI)

Heterogeneity: not applicable
24
26
100.0 %
-2.70 [ -7.76, 2.36 ]
Test for overall effect: Z = 1.05 (P = 0.30) Test for subgroup differences: Not applicable
-10
-5
10
Favours treatment
Favours control
40
Analysis 1.3. Comparison 1 Psychostimulants versus placebo, Outcome 3 Clinical response (short term) any measure/denition.
Comparison: 1 Psychostimulants versus placebo Outcome: 3 Clinical response (short term) - any measure/denition
Study or subgroup
Treatment n/N
Control n/N
Odds Ratio M-H,Fixed,95% CI
Weight
1 Psychostimulants Elizur 1979 Robin 1958 Wheatley 1969 a 6/10 10/17 26/38 3/10 13/23 25/32 8.4 % 31.8 % 59.9 % 3.50 [ 0.55, 22.30 ] 1.10 [ 0.31, 3.91 ] 0.61 [ 0.21, 1.79 ]
Subtotal (95% CI)

Total events: 42 (Treatment), 41 (Control)
65
65
100.0 %
1.01 [ 0.48, 2.09 ]
Subtotal (95% CI)

Total events: 0 (Treatment), 0 (Control) Heterogeneity: not applicable Test for overall effect: not applicable
0.0 %
0.0 [ 0.0, 0.0 ]
0.1 0.2
0.5
10
Favours treatment
Favours control
41
Analysis 1.4. Comparison 1 Psychostimulants versus placebo, Outcome 4 Improvement in cognitive functioning (short term).
Comparison: 1 Psychostimulants versus placebo Outcome: 4 Improvement in cognitive functioning (short term)
Study or subgroup
Control N Mean(SD)
Weight
1 Psychostimulants Lee 2005 Wagner 2000 10 11 32.8 (2.6) 64.2 (20.2) 10 11 31 (1.6) 61.2 (32.5) 45.3 % 54.7 % 0.80 [ -0.12, 1.72 ] 0.11 [ -0.73, 0.94 ]
Subtotal (95% CI)
21
21
100.0 %
0.42 [ -0.20, 1.04 ]
Subtotal (95% CI)

0.0 %
0.0 [ 0.0, 0.0 ]
-10
-5
10
Favours treatment
Favours control
42
Analysis 1.5. Comparison 1 Psychostimulants versus placebo, Outcome 5 Reduction in fatigue (short term).
Comparison: 1 Psychostimulants versus placebo Outcome: 5 Reduction in fatigue (short term)
Study or subgroup
Control N Mean(SD)
Weight
1 Psychostimulants Elizur 1979 Wagner 2000 10 11 3.13 (0.35) 18.3 (3.9) 10 11 3.84 (0.33) 25.4 (4.4) 44.2 % 55.8 % -2.00 [ -3.12, -0.88 ] -1.64 [ -2.64, -0.65 ]
Subtotal (95% CI)
21
21
100.0 %
-1.80 [ -2.54, -1.06 ]
Heterogeneity: Chi2 = 0.22, df = 1 (P = 0.64); I2 =0.0% Test for overall effect: Z = 4.76 (P < 0.00001) 2 Modanil
Subtotal (95% CI)

0.0 %
0.0 [ 0.0, 0.0 ]
-10
-5
10
Favours treatment
Favours control
43
Analysis 1.6. Comparison 1 Psychostimulants versus placebo, Outcome 6 Reduction in fatigue (medium term).
Comparison: 1 Psychostimulants versus placebo Outcome: 6 Reduction in fatigue (medium term)
Study or subgroup
Control N Mean(SD)
Weight
1 Psychostimulants
Subtotal (95% CI)

0.0 %
0.0 [ 0.0, 0.0 ]
22
3.52 (2.74)
19
3.28 (2.73)
100.0 %
0.24 [ -1.44, 1.92 ]
Subtotal (95% CI)

22
19
100.0 %
0.24 [ -1.44, 1.92 ]
-10
-5
10
Favours treatment
Favours control
44
Analysis 1.7. Comparison 1 Psychostimulants versus placebo, Outcome 7 Reduction in hypersomnia (medium term): Epworth Sleepiness Scale.
Comparison: 1 Psychostimulants versus placebo Outcome: 7 Reduction in hypersomnia (medium term): Epworth Sleepiness Scale
Study or subgroup
Control N Mean(SD)
Weight
1 Psychostimulants
Subtotal (95% CI)

0.0 %
0.0 [ 0.0, 0.0 ]
22
6 (4.35)
19
6.24 (4.79)
100.0 %
-0.24 [ -3.06, 2.58 ]
Subtotal (95% CI)

22
19
100.0 %
-0.24 [ -3.06, 2.58 ]
-10
-5
10
Favours treatment
Favours control
45
Analysis 1.8. Comparison 1 Psychostimulants versus placebo, Outcome 8 Improvement in health-related quality of life (short term).
Comparison: 1 Psychostimulants versus placebo Outcome: 8 Improvement in health-related quality of life (short term)
Study or subgroup
Control N Mean(SD)
Weight
1 Psychostimulants Lee 2005 Wagner 2000 10 11 119.4 (11.9) 7.9 (2.4) 10 11 109.8 (31.9) 6 (2.4) 49.2 % 50.8 % 0.38 [ -0.50, 1.27 ] 0.76 [ -0.11, 1.63 ]
Subtotal (95% CI)
21
21
100.0 %
0.57 [ -0.05, 1.20 ]
Heterogeneity: Chi2 = 0.36, df = 1 (P = 0.55); I2 =0.0% Test for overall effect: Z = 1.81 (P = 0.070) 2 Modanil
Subtotal (95% CI)

0.0 %
0.0 [ 0.0, 0.0 ]
-10
-5
10
Favours treatment
Favours control
46
Analysis 1.9. Comparison 1 Psychostimulants versus placebo, Outcome 9 Acceptability: dropouts due to side effects (short term).
Comparison: 1 Psychostimulants versus placebo Outcome: 9 Acceptability: dropouts due to side effects (short term)
Study or subgroup
Treatment n/N
Control n/N
Weight
1 Psychostimulants Robin 1958 Wagner 2000 2/22 0/11 0/23 1/12 24.0 % 76.0 % 5.73 [ 0.26, 126.42 ] 0.33 [ 0.01, 9.07 ]
Subtotal (95% CI)

33
35
100.0 %
1.63 [ 0.26, 10.25 ]
Subtotal (95% CI)

0.0 %
0.0 [ 0.0, 0.0 ]
0.1 0.2
0.5
10
Favours treatment
Favours control
47
Analysis 1.10. Comparison 1 Psychostimulants versus placebo, Outcome 10 Tolerability: proportion of participants experiencing side effects (short term).
Comparison: 1 Psychostimulants versus placebo Outcome: 10 Tolerability: proportion of participants experiencing side effects (short term)
Study or subgroup
Treatment n/N
Control n/N
Odds Ratio MH,Random,95% CI
Weight
1 Psychostimulants Rickels 1970 Rickels 1972 15/63 28/46 2/27 28/49 34.8 % 65.2 % 3.91 [ 0.83, 18.45 ] 1.17 [ 0.51, 2.65 ]
Subtotal (95% CI)
109
76
100.0 %
1.78 [ 0.57, 5.55 ]
Total events: 43 (Treatment), 30 (Control) Heterogeneity: Tau2 = 0.34; Chi2 = 1.85, df = 1 (P = 0.17); I2 =46% Test for overall effect: Z = 0.99 (P = 0.32) 2 Modanil
Subtotal (95% CI)

0.0 %
0.0 [ 0.0, 0.0 ]
0.1 0.2
0.5
10
Favours treatment
Favours control
48
Analysis 1.11. Comparison 1 Psychostimulants versus placebo, Outcome 11 Tolerability: proportion of participants experiencing side effects (medium term).
Comparison: 1 Psychostimulants versus placebo Outcome: 11 Tolerability: proportion of participants experiencing side effects (medium term)
Study or subgroup
Treatment n/N
Control n/N
Weight
1 Psychostimulants Elizur 1979 Wheatley 1969 a 3/10 18/38 1/10 3/32 23.0 % 77.0 % 3.86 [ 0.33, 45.57 ] 8.70 [ 2.26, 33.51 ]
Subtotal (95% CI)

48
42
100.0 %
7.22 [ 2.21, 23.57 ]
Heterogeneity: Tau2 = 0.0; Chi2 = 0.32, df = 1 (P = 0.57); I2 =0.0% Test for overall effect: Z = 3.27 (P = 0.0011) 2 Modanil
Subtotal (95% CI)

0.0 %
0.0 [ 0.0, 0.0 ]
0.1 0.2
0.5
10
Favours treatment
Favours control
49
Analysis 1.12. Comparison 1 Psychostimulants versus placebo, Outcome 12 Tolerability: dizziness (short term).
Comparison: 1 Psychostimulants versus placebo Outcome: 12 Tolerability: dizziness (short term)
Study or subgroup
Treatment n/N
Control n/N
Weight
Subtotal (95% CI)

114
76
100.0 %
1.31 [ 0.40, 4.24 ]
Subtotal (95% CI)

0.0 %
0.0 [ 0.0, 0.0 ]
0.1 0.2
0.5
10
Favours treatment
Favours control
50
Analysis 1.13. Comparison 1 Psychostimulants versus placebo, Outcome 13 Tolerability: drowsiness (short term).
Comparison: 1 Psychostimulants versus placebo Outcome: 13 Tolerability: drowsiness (short term)
Study or subgroup
Treatment n/N
Control n/N
Weight
Subtotal (95% CI)

114
76
100.0 %
0.41 [ 0.12, 1.37 ]
Subtotal (95% CI)

0.0 %
0.0 [ 0.0, 0.0 ]
0.1 0.2
0.5
10
Favours treatment
Favours control
51
Analysis 1.14. Comparison 1 Psychostimulants versus placebo, Outcome 14 Tolerability: headache (short term).
Comparison: 1 Psychostimulants versus placebo Outcome: 14 Tolerability: headache (short term)
Study or subgroup
Treatment n/N
Control n/N
Weight
1 Psychostimulants Elizur 1979 Rickels 1972 0/10 6/46 1/10 5/49 25.4 % 74.6 % 0.30 [ 0.01, 8.33 ] 1.32 [ 0.37, 4.66 ]
Subtotal (95% CI)

56
59
100.0 %
1.06 [ 0.34, 3.35 ]
Subtotal (95% CI)

0.0 %
0.0 [ 0.0, 0.0 ]
0.1 0.2
0.5
10
Favours treatment
Favours control
52
Analysis 1.15. Comparison 1 Psychostimulants versus placebo, Outcome 15 Tolerability: insomnia (short term).
Comparison: 1 Psychostimulants versus placebo Outcome: 15 Tolerability: insomnia (short term)
Study or subgroup
Treatment n/N
Control n/N
Weight
1 Psychostimulants Rickels 1972 11/46 7/49 100.0 % 1.89 [ 0.66, 5.38 ]
Subtotal (95% CI)

Total events: 11 (Treatment), 7 (Control) Heterogeneity: not applicable Test for overall effect: Z = 1.19 (P = 0.24) 2 Modanil
46
49
100.0 %
1.89 [ 0.66, 5.38 ]
Subtotal (95% CI)

0.0 %
0.0 [ 0.0, 0.0 ]
0.1 0.2
0.5
10
Favours treatment
Favours control
53
Analysis 1.16. Comparison 1 Psychostimulants versus placebo, Outcome 16 Tolerability: nausea (short term).
Comparison: 1 Psychostimulants versus placebo Outcome: 16 Tolerability: nausea (short term)
Study or subgroup
Treatment n/N
Control n/N
Weight
Subtotal (95% CI)

114
76
100.0 %
1.52 [ 0.31, 7.57 ]
Subtotal (95% CI)

0.0 %
0.0 [ 0.0, 0.0 ]
0.1 0.2
0.5
10
Favours treatment
Favours control
54
Analysis 1.17. Comparison 1 Psychostimulants versus placebo, Outcome 17 Tolerability: tremor (short term).
Comparison: 1 Psychostimulants versus placebo Outcome: 17 Tolerability: tremor (short term)
Study or subgroup
Treatment n/N
Control n/N
Weight
1 Psychostimulants Elizur 1979 Rickels 1972 2/10 3/46 1/10 1/49 46.9 % 53.1 % 2.25 [ 0.17, 29.77 ] 3.35 [ 0.34, 33.41 ]
Subtotal (95% CI)

56
59
100.0 %
2.83 [ 0.51, 15.67 ]
Subtotal (95% CI)

0.0 %
0.0 [ 0.0, 0.0 ]
0.1 0.2
0.5
10
Favours treatment
Favours control
55
Analysis 2.1. Comparison 2 Psychostimulants versus antidepressants, Outcome 1 Reduction in depression symptoms (short term).
Comparison: 2 Psychostimulants versus antidepressants Outcome: 1 Reduction in depression symptoms (short term)
Study or subgroup
Control N Mean(SD)
Weight
1 Psychostimulants Lee 2005 10 15.7 (5.6) 10 20 (4.6) 100.0 % -4.30 [ -8.79, 0.19 ]
Subtotal (95% CI)

10
10
100.0 %
-4.30 [ -8.79, 0.19 ]
Test for overall effect: Z = 1.88 (P = 0.061) 2 Modanil
Subtotal (95% CI)

0.0 %
0.0 [ 0.0, 0.0 ]
-10
-5
10
Favours treatment
Favours control
56
Analysis 2.2. Comparison 2 Psychostimulants versus antidepressants, Outcome 2 Tolerability: number of participants experiencing side effects (medium term).
Comparison: 2 Psychostimulants versus antidepressants Outcome: 2 Tolerability: number of participants experiencing side effects (medium term)
Study or subgroup
Treatment n/N
Control n/N
Weight
1 Psychostimulants Fernandez 1995 5/6 7/9 100.0 % 1.43 [ 0.10, 20.44 ]
Subtotal (95% CI)

100.0 %
1.43 [ 0.10, 20.44 ]
Subtotal (95% CI)

0.0 %
0.0 [ 0.0, 0.0 ]
0.1 0.2
0.5
10
Favours treatment
Favours control
57
Analysis 3.1. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 1 Reduction in depression symptoms (short term).
Comparison: 3 Psychostimulants versus placebo as adjunct to antidepressant treatment Outcome: 1 Reduction in depression symptoms (short term)
Study or subgroup
Control N Mean(SD)
Weight
1 Psychostimulants Patkar 2006 28 12.5 (7.7) 22 14.1 (7.9) 100.0 % -1.60 [ -5.96, 2.76 ]
Subtotal (95% CI)

28
22
100.0 %
-1.60 [ -5.96, 2.76 ]
Test for overall effect: Z = 0.72 (P = 0.47) 2 Modanil DeBattista 2003 Fava 2005 65 143 13.43 (4.1) 9.8 (5.66) 63 140 12.94 (4.7) 10.3 (5.33) 41.2 % 58.8 % 0.49 [ -1.04, 2.02 ] -0.50 [ -1.78, 0.78 ]
Subtotal (95% CI)
208
203
100.0 %
-0.09 [ -1.07, 0.89 ]
Heterogeneity: Chi2 = 0.95, df = 1 (P = 0.33); I2 =0.0% Test for overall effect: Z = 0.18 (P = 0.85) Test for subgroup differences: Chi2 = 0.44, df = 1 (P = 0.51), I2 =0.0%
-10
-5
10
Favours treatment
Favours control
58
Analysis 3.2. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 2 Reduction in depression symptoms (medium term).
Comparison: 3 Psychostimulants versus placebo as adjunct to antidepressant treatment Outcome: 2 Reduction in depression symptoms (medium term)
Study or subgroup
Control N Mean(SD)
Weight
1 Psychostimulants
Subtotal (95% CI)

Heterogeneity: not applicable Test for overall effect: not applicable 2 Modanil DeBattista 2003 Fava 2005
0.0 %
0.0 [ 0.0, 0.0 ]
68 156
12.26 (5.7) 8.8 (5.61)
67 152
12.6 (5.9) 9.8 (5.96)
30.4 % 69.6 %
-0.34 [ -2.30, 1.62 ] -1.00 [ -2.29, 0.29 ]
Subtotal (95% CI)
224
219
100.0 %
-0.80 [ -1.88, 0.28 ]
Heterogeneity: Chi2 = 0.30, df = 1 (P = 0.58); I2 =0.0% Test for overall effect: Z = 1.45 (P = 0.15) Test for subgroup differences: Not applicable
-10
-5
10
Favours treatment
Favours control
59
Analysis 3.3. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 3 Clinical response (short term).
Comparison: 3 Psychostimulants versus placebo as adjunct to antidepressant treatment Outcome: 3 Clinical response (short term)
Study or subgroup
Treatment n/N
Control n/N
Weight
1 Psychostimulants Patkar 2006 12/27 7/23 100.0 % 1.83 [ 0.57, 5.88 ]
Subtotal (95% CI)

27
23
100.0 %
1.83 [ 0.57, 5.88 ]
Subtotal (95% CI)

0.0 %
0.0 [ 0.0, 0.0 ]
0.1 0.2
0.5
10
Favours treatment
Favours control
60
Analysis 3.4. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 4 Reduction in fatigue symptoms (short term).
Comparison: 3 Psychostimulants versus placebo as adjunct to antidepressant treatment Outcome: 4 Reduction in fatigue symptoms (short term)
Study or subgroup
Control N Mean(SD)
Mean Difference IV,Random,95% CI
Weight
Mean Difference IV,Random,95% CI
1 Psychostimulants
Subtotal (95% CI)

0.0 %
0.0 [ 0.0, 0.0 ]
63 156
5.04 (1.2) 6.8 (2.3)
66 152
4.59 (1.4) 7.2 (2.5)
51.6 % 48.4 %
0.45 [ 0.00, 0.90 ] -0.40 [ -0.94, 0.14 ]
Subtotal (95% CI)
219
218
100.0 %
0.04 [ -0.79, 0.87 ]
Heterogeneity: Tau2 = 0.30; Chi2 = 5.66, df = 1 (P = 0.02); I2 =82% Test for overall effect: Z = 0.09 (P = 0.93)
-10
-5
10
Favours treatment
Favours control
61
Analysis 3.5. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 5 Reduction in fatigue symptoms (medium term).
Comparison: 3 Psychostimulants versus placebo as adjunct to antidepressant treatment Outcome: 5 Reduction in fatigue symptoms (medium term)
Study or subgroup
Control N Mean(SD)
Weight
1 Psychostimulants
Subtotal (95% CI)

0.0 %
0.0 [ 0.0, 0.0 ]
68 156
4.63 (1.3) 6.5 (2.6)
67 152
4.73 (1.5) 7.1 (2.4)
58.2 % 41.8 %
-0.10 [ -0.57, 0.37 ] -0.60 [ -1.16, -0.04 ]
Subtotal (95% CI)
224
219
100.0 %
-0.31 [ -0.67, 0.05 ]
Heterogeneity: Chi2 = 1.79, df = 1 (P = 0.18); I2 =44% Test for overall effect: Z = 1.68 (P = 0.093) Test for subgroup differences: Not applicable
-10
-5
10
Favours treatment
Favours control
62
Analysis 3.6. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 6 Reduction in hypersomnia (short term): Epworth Sleepiness Scale.
Comparison: 3 Psychostimulants versus placebo as adjunct to antidepressant treatment Outcome: 6 Reduction in hypersomnia (short term): Epworth Sleepiness Scale
Study or subgroup
Control N Mean(SD)
Weight
1 Psychostimulants
Subtotal (95% CI)

0.0 %
0.0 [ 0.0, 0.0 ]
66 156
7.85 (4.4) 11.2 (4.4)
63 152
8.6 (4.6) 12 (4.6)
29.5 % 70.5 %
-0.75 [ -2.30, 0.80 ] -0.80 [ -1.81, 0.21 ]
Subtotal (95% CI)
222
215
100.0 %
-0.79 [ -1.63, 0.06 ]
-10
-5
10
Favours treatment
Favours control
63
Analysis 3.7. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 7 Reduction in hypersomnia (medium term): Epworth Sleepiness Scale.
Comparison: 3 Psychostimulants versus placebo as adjunct to antidepressant treatment Outcome: 7 Reduction in hypersomnia (medium term): Epworth Sleepiness Scale
Study or subgroup
Control N Mean(SD)
Weight
1 Psychostimulants
Subtotal (95% CI)

0.0 %
0.0 [ 0.0, 0.0 ]
68 156
7 (3.9) 10.5 (4.7)
67 152
8.82 (5.1) 11.5 (4.8)
32.4 % 67.6 %
-1.82 [ -3.35, -0.29 ] -1.00 [ -2.06, 0.06 ]
Subtotal (95% CI)
224
219
100.0 %
-1.27 [ -2.14, -0.39 ]
-10
-5
10
Favours treatment
Favours control
64
Analysis 3.8. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 8 Acceptability: dropouts due to side effects (short term).
Comparison: 3 Psychostimulants versus placebo as adjunct to antidepressant treatment Outcome: 8 Acceptability: dropouts due to side effects (short term)
Study or subgroup
Treatment n/N
Control n/N
Weight
Subtotal (95% CI)

30
30
100.0 %
1.00 [ 0.13, 7.60 ]
Subtotal (95% CI)

0.0 %
0.0 [ 0.0, 0.0 ]
0.1 0.2
0.5
10
Favours treatment
Favours control
65
Analysis 3.9. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 9 Acceptability: dropouts due to side effects (medium term).
Comparison: 3 Psychostimulants versus placebo as adjunct to antidepressant treatment Outcome: 9 Acceptability: dropouts due to side effects (medium term)
Study or subgroup
Treatment n/N
Control n/N
Weight
1 Psychostimulants Postolache 1999 2/5 0/4 100.0 % 6.43 [ 0.23, 181.82 ]
Subtotal (95% CI)

Total events: 2 (Treatment), 0 (Control) Heterogeneity: not applicable Test for overall effect: Z = 1.09 (P = 0.28) 2 Modanil DeBattista 2003 Fava 2005
100.0 %
6.43 [ 0.23, 181.82 ]
3/69 9/159
4/67 5/155
44.8 % 55.2 %
0.72 [ 0.15, 3.33 ] 1.80 [ 0.59, 5.50 ]
Subtotal (95% CI)
228
222
100.0 %
1.31 [ 0.54, 3.18 ]
Total events: 12 (Treatment), 9 (Control) Heterogeneity: Chi2 = 0.91, df = 1 (P = 0.34); I2 =0.0% Test for overall effect: Z = 0.61 (P = 0.54)
0.1 0.2
0.5
10
Favours treatment
Favours control
66
Analysis 3.10. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 10 Acceptability: dropouts due to lack of efcacy (medium term).
Comparison: 3 Psychostimulants versus placebo as adjunct to antidepressant treatment Outcome: 10 Acceptability: dropouts due to lack of efcacy (medium term)
Study or subgroup
Treatment n/N
Control n/N
Weight
1 Psychostimulants
Subtotal (95% CI)

Total events: 0 (Treatment), 0 (Control) Heterogeneity: not applicable Test for overall effect: not applicable 2 Modanil DeBattista 2003 Fava 2005
0.0 %
0.0 [ 0.0, 0.0 ]
3/69 1/159
2/67 2/155
49.1 % 50.9 %
1.48 [ 0.24, 9.13 ] 0.48 [ 0.04, 5.39 ]
Subtotal (95% CI)

228
222
100.0 %
0.97 [ 0.24, 3.96 ]
Heterogeneity: Chi2 = 0.52, df = 1 (P = 0.47); I2 =0.0% Test for overall effect: Z = 0.04 (P = 0.97)
0.1 0.2
0.5
10
Favours treatment
Favours control
67
Analysis 3.11. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 11 Tolerability: proportion of participants experiencing side effects (short term).
Comparison: 3 Psychostimulants versus placebo as adjunct to antidepressant treatment Outcome: 11 Tolerability: proportion of participants experiencing side effects (short term)
Study or subgroup
Treatment n/N
Control n/N
Weight
Subtotal (95% CI)

30
30
100.0 %
1.32 [ 0.47, 3.72 ]
Subtotal (95% CI)

0.0 %
0.0 [ 0.0, 0.0 ]
0.1 0.2
0.5
10
Favours treatment
Favours control
68
Analysis 3.12. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 12 Tolerability: headache (medium term).
Comparison: 3 Psychostimulants versus placebo as adjunct to antidepressant treatment Outcome: 12 Tolerability: headache (medium term)
Study or subgroup
Treatment n/N
Control n/N
Weight
1 Psychostimulants
Subtotal (95% CI)

0.0 %
0.0 [ 0.0, 0.0 ]
15/69 21/158
8/67 24/153
42.6 % 57.4 %
2.05 [ 0.80, 5.21 ] 0.82 [ 0.44, 1.55 ]
Subtotal (95% CI)
227
220
100.0 %
1.21 [ 0.50, 2.94 ]
Total events: 36 (Treatment), 32 (Control) Heterogeneity: Tau2 = 0.25; Chi2 = 2.50, df = 1 (P = 0.11); I2 =60% Test for overall effect: Z = 0.43 (P = 0.67)
0.1 0.2
0.5
10
Favours treatment
Favours control
69
Analysis 3.13. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 13 Tolerability: insomnia (short term).
Comparison: 3 Psychostimulants versus placebo as adjunct to antidepressant treatment Outcome: 13 Tolerability: insomnia (short term)
Study or subgroup
Treatment n/N
Control n/N
Weight
Subtotal (95% CI)

30
30
100.0 %
1.00 [ 0.19, 5.40 ]
Subtotal (95% CI)

0.0 %
0.0 [ 0.0, 0.0 ]
0.1 0.2
0.5
10
Favours treatment
Favours control
70
Analysis 3.14. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 14 Tolerability: insomnia (medium term).
Comparison: 3 Psychostimulants versus placebo as adjunct to antidepressant treatment Outcome: 14 Tolerability: insomnia (medium term)
Study or subgroup
Treatment n/N
Control n/N
Weight
1 Psychostimulants
Subtotal (95% CI)

0.0 %
0.0 [ 0.0, 0.0 ]
13/69 7/158
9/67 7/153
52.2 % 47.8 %
1.50 [ 0.59, 3.78 ] 0.97 [ 0.33, 2.82 ]
Subtotal (95% CI)
227
220
100.0 %
1.24 [ 0.62, 2.50 ]
0.1 0.2
0.5
10
Favours treatment
Favours control
71
Analysis 3.15. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 15 Tolerability: nausea (short term).
Comparison: 3 Psychostimulants versus placebo as adjunct to antidepressant treatment Outcome: 15 Tolerability: nausea (short term)
Study or subgroup
Treatment n/N
Control n/N
Weight
Subtotal (95% CI)

30
30
100.0 %
3.22 [ 0.32, 32.89 ]
Subtotal (95% CI)

0.0 %
0.0 [ 0.0, 0.0 ]
0.1 0.2
0.5
10
Favours treatment
Favours control
72
Analysis 3.16. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 16 Tolerability: nausea (medium term).
Comparison: 3 Psychostimulants versus placebo as adjunct to antidepressant treatment Outcome: 16 Tolerability: nausea (medium term)
Study or subgroup
Treatment n/N
Control n/N
Weight
1 Psychostimulants
Subtotal (95% CI)

0.0 %
0.0 [ 0.0, 0.0 ]
3/69 15/158
5/67 3/153
48.5 % 51.5 %
0.56 [ 0.13, 2.46 ] 5.24 [ 1.49, 18.50 ]
Subtotal (95% CI)
227
220
100.0 %
1.78 [ 0.20, 15.95 ]
Total events: 18 (Treatment), 8 (Control) Heterogeneity: Tau2 = 2.02; Chi2 = 5.13, df = 1 (P = 0.02); I2 =81% Test for overall effect: Z = 0.51 (P = 0.61)
0.1 0.2
0.5
10
Favours treatment
Favours control
73
Analysis 3.17. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 17 8Tolerability: tremor (short term).
Comparison: 3 Psychostimulants versus placebo as adjunct to antidepressant treatment Outcome: 17 8Tolerability: tremor (short term)
Study or subgroup
Treatment n/N
Control n/N
Weight
Subtotal (95% CI)

30
30
100.0 %
0.48 [ 0.04, 5.63 ]
Subtotal (95% CI)

0.0 %
0.0 [ 0.0, 0.0 ]
0.1 0.2
0.5
10
Favours treatment
Favours control
74
Analysis 3.18. Comparison 3 Psychostimulants versus placebo as adjunct to antidepressant treatment, Outcome 18 Tolerability: upper respiratory tract infection (medium term).
Comparison: 3 Psychostimulants versus placebo as adjunct to antidepressant treatment Outcome: 18 Tolerability: upper respiratory tract infection (medium term)
Study or subgroup
Treatment n/N
Control n/N
Weight
1 Psychostimulants
Subtotal (95% CI)

0.0 %
0.0 [ 0.0, 0.0 ]
5/69 5/158
4/67 9/153
29.8 % 70.2 %
1.23 [ 0.32, 4.79 ] 0.52 [ 0.17, 1.60 ]
Subtotal (95% CI)
227
220
100.0 %
0.73 [ 0.31, 1.71 ]
0.1 0.2
0.5
10
Favours treatment
Favours control
75
Analysis 4.1. Comparison 4 Psychostimulants versus placebo: concomitant serious medical condition, Outcome 1 Reduction in depression symptoms (short term).
Comparison: 4 Psychostimulants versus placebo: concomitant serious medical condition Outcome: 1 Reduction in depression symptoms (short term)
Study or subgroup
Treatment N Mean(SD) 15.7 (5.6) 7.5 (3.9)
Control N 10 11 Mean(SD) 22.3 (4.2) 9.4 (6.1)
Std. Mean Difference IV,Fixed,95% CI -1.28 [ -2.26, -0.30 ] -0.36 [ -1.20, 0.49 ]
Lee 2005 Wagner 2000
10 11
Subtotal (95% CI)
0.0 [ 0.0, 0.0 ]
Heterogeneity: Chi2 = 0.0, df = 0 (P<0.00001); I2 =0.0% Test for overall effect: Z = 0.0 (P < 0.00001) Test for subgroup differences: Not applicable
-4
-2
Favours treatment
Favours control
ADDITIONAL TABLES
Table 1. Methodological quality of included trials
Trial 1-Cantello 1989
Sample size
Recruitment rate
Randomisation
Masking
Attrition
No sample size cal- No details provided culation
States randomised Double-blind: No loss to follow up sequence No details who is masked No details provided Double-blind: No loss to follow up No details who is masked Double-blind: 14% No details who is of 69 in modanil masked group, 12% of 67 in placebo group, unclear if used intention-to-treat analysis Double-blind: No loss to follow up No details who is masked
2-Cookson 1986
3-DeBattista 2003
No details provided
4-Elizur 1979
No details provided
76
(Continued)
5-Fava 2005
Unclear
Unclear
ComDouble-blind: puter generated ran- All site personnel dom numbers blinded except the research pharmacist
16% of 155 in placebo group, 15% of 159 in intervention group lost to follow up. No Intention to treat analysis
6-Fernandez 1995
No details provided
Double-blind: 25% of 12 No details who is in methylphenidate masked group, 25% of 8 in desipamine group lost to follow up. No Intention to treat analysis Double-blind: Pa- 1.5% of 46 lost to tient and interven- follow up. No intentionist tion to treat analysis Double-blind: 26% of 106 lost to No details who is follow up. No intenmasked tion to treat analysis Double-blind: No loss to follow-up No details who is masked Double-blind: No loss to follow-up No details who is masked Double-blind: 5% of 20 lost to folNo details who is low-up. No intenmasked tion-to-treat analysis
7-Hare 1962
No sample size cal- No details provdied culation
No details provided
8-Hare 1964
No details provided
9-Kits 1970
No details provided
10-Lee 2005
Sample size calcu- No details provided lation reported but not met No sample size cal- No details provided culation
No details provided
11-Mattes 1985
No details provided
12-Overall 1962
Randomised blocks Double-blind: 41% of 204 lost to - patients assigned No details who is folin order of admis- masked low-up. No intension tion-to-treat analysis No details provided Double-blind: 17% No details who is of 60 lost to follow masked up. Used intentionto-treat analysis
13-Patkar 2006
77
(Continued)
14-Postolache 1999 No sample size cal- No details provided culation
No details provided
Double-blind: 22% of 9 lost to folNo details who is low-up. No intenmasked tion-to-treat analysis. Double-blind: Pa- No loss to follow-up tients and assessor
15--Reus 1979
No sample size cal- No details provided culation No sample size cal- No details provided culation
No details provided
16-Rickels 1970
PaDouble-blind: 31% of placebo tients systematically No details who is group, 16% of perandomly assigned masked moline group, 16% in sucessive nummethylphenidate bers within each of group lost to follow the three healthcare up. No Intention to settings treat analysis Systematised ran- Double-blind: Unclear in paper dom order shufed No details who is masked Sealed envelopes Double-blind: were prepared Patient and clinican containing the prescription of placebo or methylphenidate and were throughly shufed No details provided 23% of 22 in intervention group lost to follow up, no loss to follow-up in control group
17-Rickels 1972
18-Robin 1958
19-Silberman 1981
Double-blind. Pa- No loss to follow up tient masked, no details on who else is masked
20-Vaishnavi 2006
RanDouble-blind: 8% dom numbers cho- No details who is of 24 in modanil sen from a statisti- masked group and 27% in cal distribution with placebo group lost a computer program to follow-up. Intenand each patient astion-to-treat analysigned a number. sis Patients who recieved a higher number from the computer placed in one group, all remaining in another
78
(Continued)
21-Wagner 2000
No sample size cal- Indirect No details provided culation recruitment through newsletters and posters No sample size cal- No details provided culation No details provided
Double-blind: 4% of 23 lost to folNo details who is low-up. No intenmasked tion-to-treat analysis Double-blind: Patient and assessor 19% of 16 lost to followup. No intentionto-treat analysis
22-Wallace 1995
23-Wheatley 1969a No sample size cal- No details provided culation 24-Wheatley 1969b No sample size cal- No details provided culation
No details provided
Double blind: No No loss to follow up details provided Double blind: No No loss to follow up details provided
No details provided
Table 2. Other symptoms (other than depressed mood) evaluated in one trial only
Symptom (Reference) Alertness (Robin 1958) Anorexia (Elizur 1979) Appetite (Cookson 1986)
Participants 40 20 20
Outcome effect OR 1.43, 95% CI 0.45, 5.50 WMD 0.97, 95% CI 0.52,1.42 In the responder (dened as a fall in depression rating greater after PS than placebo) group all experienced an increase in appetite following PS, which was generally greater than that experienced after placebo. In the non-responders 1/ 10 showed a signicant elevation of hunger after PS. In the four where response was eqivocal there was litle or no change in appetite WMD -0.21, 95% CI 0.53,0.11 OR 1.58, 95% CI 0.45, 5.50 OR 1.44, 95% CI 1.08, 1,80 WMD 0.17, 95% CI -0.22,0.56 WMD -0.18, 95% CI -0.39, 0.03 OR 1.23, 95% CI 0.23, 4.31 OR 0.78, 95% CI 0.20, 3.02 WMD -0.17, 95% CI -0.54, 0.20
Concentration (Elizur 1979) Condence (Robin 1958) Insomnia (Elizur 1979) Libido (Elizur 1979) Memory (Elizur 1979) Range of interests (Robin 1958) Retardation (Robin 1958) Tension (Elizur 1979)
20 40 20 20 20 40 40 20
79
Table 3. Side effects assessed in one trial only Side effect Sample size Effect outcome
PS VERSUS PLACEBO Anxiety Constipation Dizziness Dry mouth Excitement Headache Loss of appetite Nausea Palpitation Restlessness Tachycardia Tremor Tolerance Vertigo Weight gain Weight loss PS VERSUS ANTIDEPRESSANT Headache Insomnia Nausea MODAFINIL VERSUS PLACEBO Anxiety 136 OR 1.23, 95% 0.32, 4.79
80
60 95 20 55 95 20 60 60 20 20 95 20 20 95 95 95
OR 2.07, 95% CI 0.18, 24.15 OR 2.30, 95% CI 0.54, 9.80 OR 2.25, 95% CI 0.17, 29.77 OR 3.23, 95% CI 0.80, 13.20 OR 1.25, 95% CI 0.44, 3.57 OR 0.30, 95% 0.01, 8.33 OR 3.22, 95% CI 0.23, 32.89 OR 3.22, 95% CI 0.32, 32.89 OR 0.30, 95% CI 0.01, 8.33 OR 1.00, 95% CI 0.05, 18.57 OR 1.25, 95% CI 0.44, 3.57 OR 2.25, 95% 0.17, 29.77 Reports no addictive tendency was observed (No data provided) OR 1.22, 95% CI 0.05, 30.03 OR 0.27, 95% CI 0.05, 1.39 OR 3.53, 95% CI 0.67, 18.45
15 15 15
OR 0.23, 95% CI 0.01, 5.73 OR 0.25, 95% CI 0.02, 3.10 OR 4.00, 95% 0.27, 58.56
Table 3. Side effects assessed in one trial only
(Continued)
Asthenia Cardiovascular
136 136; 41
OR 0.37, 95% CI 0.07, 1.98 DeBattista 2003 reported no signicant differences in vital signs, including sitting and standing systolic and diastolic blood presure and heart rate (no data provided). Vaishnavi 2006 reported there were no signicant differences in pulse (p=0.10), systolic blood pressure (p=0.38) or diastolic blood pressure (p=0.60) () (no further data provided) OR 0.56, 95% CI 0.20, 1.59 OR 0.97, 95% CI 0.27, 3.51 OR 2.00, 95% CI 0.67, 5.99 OR 0.54, 95% CI 0.16, 1.89 OR 11.52, 95% CI 0.62, 212.40 OR 0.83, 95% CI 0.28, 2.44 OR 0.23, 95% CI 0.03, 2.13 OR 0.97, 95% CI 0.27, 3.41 OR 5.43, 95% CI 1.48, 19.89 OR 1.78, 95% CI 0.50, 6.38
Diarrhoea -short term Diarrhoea - medium term Dry mouth Hypertension Hypertonia Infection Myalgia Nasophartngitis Nervousness Rhinitis
311 136 311 311 136 136 136 311 136 136
Table 4. Outcomes on depressed mood in trials not included in the meta-analysis
Trial Cantello 1989
Trial details
Effect outcome
Direction of outcome
Depression measured by observers In group of 13 depressed Parkinsoni- Mixed global judgement score. PS: Methy- ans no signicant difference in selflamphetamine rated and observer judgement score between PS and placebo. In 14 patients with major depression they was signicant relief of depressive symptoms when given PS but not placebo. (pre-cross over outcomes not reported separately) Self-rated 10cm VAS for depression. PS 7/21 patients showed a fall ( No difference PS: Methylamphetaime >20mm) in self-rated and observer rated depression which was at least
81
Cookson 1986
(Continued)
15mm greater than the changes occuring after placebo. Four patients experienced a similar improvement in mood following both PS and placebo. Ten patients showed no mood elevation in response to either PS or placebo Hare 1962 Depression measured by clinican No signicant difference between PS No difference on a 5-point scale. PS: Dexam- and placebo. phetamine Depression measured by clinican No signicant difference between PS No difference on a 5-point scale. PS: Dexam- and antidepressant. phetamine. Ajunct to amylobarbitone Depression measured by vital syn- A difference in favour of PS was ap- No difference drome interview. PS: Methylam- parent phetamine Depression measured by psychiatrist No evidence of benet effect be- No difference completed SADS-C, self completed tween PS and placebo POMS and patient and psychiatrist completed 8-point scales of global improvement. PS: Methylphenidate Depression measured by specically No signicant difference between PS No difference derived 31 item depression scale and comparison treatments from IMPS. PS: Dexamphetamine. Ajunct to amylobarbitone Self-rated depression. PS: Dex- After PS 6/9 reported a better re- Positive troamphetamine sponse on PS than on placebo Depression measured by Zung De- PS produced signicantly more im- Unclear pression Scale. PS: Pemoline and provement than placebo in clinic Methylphenidate and general practice patients. No signicant difference in private psychiatric patients was observed between PS and placebo Depression measured by Zung De- The mean Zung score was signi- Positive pression Scale. PS: Methylphenidate cantly lower inthe PS group than in the placebo group (p<0.05) Depression measured by physi- Mean change score in depressive No difference cans BPRS ratings. PS: Dextroam- mood was not signicant after PS or phetamine after placebo
82
Hare 1964
Kits 1970
Mattes 1985
Overall 1962
Reus 1979
Rickels 1970
Rickels 1972
Silberman 1981
(Continued)
Wallace 1995
Depression measured Post hoc multiple comparisons re- No difference by Hamilton Depression Scale. PS: vealed that for 10 subjects who were Methylphenidate given the active drug rst there was no signicant difference between PS and placebo. For the three subjects who recieved placebo rst thare was no signicant difference GP rated depression. PS: Dextroam- Response to PS (44% improvement) No difference phetamine very similar to response to placebo (45%)
Wheatley 1969b
WHATS NEW
Last assessed as up-to-date: 10 February 2008.
Date 5 November 2008
Event Amended
Description Converted to new review format.
HISTORY
Protocol rst published: Issue 3, 2007 Review rst published: Issue 2, 2008
Date 11 February 2008
Event New citation required and conclusions have changed
Description Substantive amendment
CONTRIBUTIONS OF AUTHORS
BC together with MK: writing of the protocol. BC: data searches and together with LJ data selection and extraction. BC writing the review. RW: providing advice and support for statistical analysis and commentary on the ndings. AK: conceptualised topic area, advised on protocol development, commented on ndings and conclusions. LJ, MK: methodology advise and commentary on the ndings and conclusions.
83
DECLARATIONS OF INTEREST
None known.
SOURCES OF SUPPORT Internal sources

Marie Curie Palliative Care Research Unit, London, UK. Royal Free and University College Medical School, London, UK.
External sources
No sources of support supplied
INDEX TERMS Medical Subject Headings (MeSH)

Antidepressive Agents [ therapeutic use]; Central Nervous System Stimulants [ therapeutic use]; Depression [ drug therapy]; Randomized Controlled Trials as Topic
MeSH check words

Humans
84

Psicoestimulantes para Depreion

Uploaded by

Copyright:

Available Formats

Psicoestimulantes para Depreion

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Psicoestimulantes para Depreion

Uploaded by

Copyright:

Available Formats

Psychostimulants for depression (Review)

Candy B, Jones L, Williams R, Tookman A, King M

Psychostimulants for depression

Criteria for considering studies for this review

Search methods for identication of studies

Data collection and analysis

Risk of bias in included studies

construct a meaningful funnel plot.

AUTHORS CONCLUSIONS Implications for practice

Implications for research

References to studies included in this review

References to studies excluded from this review

Characteristics of included studies [ordered by study ID]

Authors judgement Unclear

Notes Risk of bias

Item Allocation concealment? DeBattista 2003 Methods Participants

Authors judgement Unclear

Authors judgement Unclear

Authors judgement Unclear

Authors judgement Unclear

Authors judgement Unclear

Description D - Not used

Authors judgement Unclear

Outcomes Notes Risk of bias Item Allocation concealment?

Authors judgement Unclear

Authors judgement Unclear

Authors judgement Unclear

Outcomes Notes Risk of bias Item Allocation concealment?

Authors judgement Unclear

Authors judgement Unclear

Authors judgement Unclear

Outcomes Notes Risk of bias Item Allocation concealment?

Authors judgement Unclear

Outcomes Notes Risk of bias Item Allocation concealment?

Authors judgement Unclear

Characteristics of excluded studies [ordered by study ID]

DATA AND ANALYSES

Comparison 1. Psychostimulants versus placebo

Statistical method Std. Mean Difference (IV, Fixed, 95% CI)

1.78 [0.57, 5.55] Not estimable

Odds Ratio (M-H, Random, 95% CI)

Comparison 2. Psychostimulants versus antidepressants

Statistical method Mean Difference (IV, Fixed, 95% CI)

1.43 [0.10, 20.44] Not estimable

Comparison 3. Psychostimulants versus placebo as adjunct to antidepressant treatment

Statistical method Mean Difference (IV, Fixed, 95% CI)

Not estimable -0.79 [-1.63, 0.06] Subtotals only

1.32 [0.47, 3.72] Not estimable

Comparison 4. Psychostimulants versus placebo: concomitant serious medical condition

Outcome or subgroup title 1 Reduction in depression symptoms (short term)

Statistical method Std. Mean Difference (IV, Fixed, 95% CI)

Effect size Subtotals only

Comparison: 1 Psychostimulants versus placebo Outcome: 1 Reduction in depression symptoms (short-term)

Std. Mean Difference IV,Fixed,95% CI

Std. Mean Difference IV,Fixed,95% CI

Subtotal (95% CI)

-0.87 [ -1.40, -0.33 ]

Subtotal (95% CI)

0.0 [ 0.0, 0.0 ]