Maths.bangor.ac.uk

Antidepressants and suicide:
risk-benefit conundrums
David Healy, MD; Chris Whitaker, MSc
Healy - Department of Psychological Medicine, University of Wales College of Medicine, Hergest Unit; Whitaker -School of Informatics, University of Wales Bangor, Bangor, United Kingdom. There has been a long-standing controversy about the possibility that selective serotonin reuptakeinhibitor (SSRI) antidepressants might induce suicidality in some patients. To shed light on this issue, thispaper reviews available randomized controlled trials (RCTs), meta-analyses of clinical trials and epidem-iological studies that have been undertaken to investigate the issue further. The original clinical studiesraising concerns about SSRIs and suicide induction produced evidence of a dose-dependent link on a chal-lenge-dechallenge and rechallenge basis between SSRIs and both agitation and suicidality. Meta-analyses ofRCTs conducted around this time indicated that SSRIs may reduce suicidal ideation in some patients.
These same RCTs, however, revealed an excess of suicidal acts on active treatments compared withplacebo, with an odds ratio of 2.4 (95% confidence interval 1.6-3.7). This excess of suicidal acts alsoappears in epidemiological studies. The data reviewed here make it difficult to sustain a null hypothesisthat SSRIs do not cause problems in some individuals. Further studies or further access to data are indi-cated to establish the magnitude of any risk and the characteristics of patients who may be most at risk.
La possibilité que les antidépresseurs inhibiteurs spécifiques du recaptage de la sérotonine (ISRS) entraînentdes tendances suicidaires chez certains patients soulève la controverse depuis longtemps. En vue d'éclairer laquestion, cette communication passe en revue les études contrôlées et randomisées (ECR), méta-analysesd'études cliniques et études épidémiologiques disponibles que l'on a effectuées pour l'approfondir. Lespremières études cliniques ayant soulevé des préoccupations au sujet des ISRS et des tendances suicidairesont produit des données probantes établissant un lien lié à la dose entre les ISRS et deux symptômes, l'agita-tion et les tendances suicidaires, suivant l'administration du médicament, la cessation, puis la reprise du traite-ment. Des méta-analyses des ECR réalisées à l'époque ont indiqué que les ISRS pourraient réduire les idéessuicidaires chez certains patients. Ces mêmes ECR ont néanmoins révélé des actes suicidaires en surnombredans le groupe de traitement actif, par rapport au placebo. À cet égard, le coefficient de probabilité s'est établià 2,4 (intervalle de confiance à 95 %, 1,6-3,7). Des études épidémiologiques ont aussi révélé des actessuicidaires en surnombre. Compte tenu des données examinées, il est difficile de soutenir l'hypothèse nullevoulant que les ISRS n'entraînent pas de problèmes chez certaines personnes. La réalisation de nouvellesétudes ou la diffusion d'autres données sont indiquées pour établir l'importance du risque et les caractéris-tiques des patients susceptibles d'être les plus vulnérables.
Correspondence to: Dr. David Healy, Department of Psychological Medicine, University of Wales College of Medicine, Hergest Unit,
Bangor, LL57 2PW United Kingdom; fax 44 1248 371397; Healy_Hergest@compuserve.com
Medical subject headings: antidepressive agents; epidemiology; meta-analysis; serotonin uptake inhibitors; suicide; treatment outcome.
J Psychiatry Neurosci 2003;28(5):331-7.
Submitted Oct. 17, 2001Revised Apr. 4, 2003; May 30, 2003Accepted June 3, 2003 J Psychiatry Neurosci 2003;28(5)
Introduction
analyzed the figures in terms of PEY only. Khan et al10found an excess of suicidal acts by individuals taking The debate regarding selective serotonin reuptake antidepressants compared with placebo, and this was inhibitors (SSRIs) and suicide started in 1990, when also replicated in another analysis,11 but the rates of sui- Teicher, Glod and Cole1 described 6 cases in which cidal acts in patients taking antidepressants and those intense suicidal preoccupation emerged during fluox- taking placebo were not significantly different in these etine treatment. This paper was followed by others,2-6 analyses. Yet, another study12 reported that rates of sui- which, combined, provided evidence of dose-response, cidal acts of patients taking antidepressants for longer challenge, dechallenge and rechallenge relations, as durations may, in fact, fall relative to placebo, which well as the emergence of an agreed mechanism by might be expected because longer term studies will which the effects were mediated and demonstrations select patients suited to the agent being investigated.
that interventions in the process could ameliorate the Although an analysis in terms of PEYs may be ap- problems. A subsequent series of reports on the effects propriate for an assessment of the risk of exposure to of sertraline and paroxetine on suicidality and akathisia placebo, it is inappropriate for the assessment of a pointed to SSRI-induced suicidality being a class effect problem that clinical studies had clearly linked to the rather than something confined to fluoxetine.7 first weeks of active therapy. An analysis of suicidal An induction of suicidality by SSRIs, therefore, had acts on the basis of duration of exposure systematically apparently been convincingly demonstrated according selects patients who do not have the problem under in- to conventional criteria for establishing cause and effect vestigation, because those with the problem often drop relations between drugs and adverse events, as laid out out of the trial, whereas others who do well are kept on by clinical trial methodologists, company investigators, treatment for months or more on grounds of compas- medico-legal authorities and the federal courts.8 Far less consistent evidence led the Medicines Control Agency The data presented by Khan and colleagues10 has in Britain in 1988 to state unambiguously that benzodi- accordingly been modified here in 4 respects (Table 1).
First, suicides and suicidal acts are presented in terms Specifically designed randomized controlled trials of absolute numbers of patients. Second, on the basis of (RCTs) on depression-related suicidality at this time an FDA paroxetine safety review13 and FDA statistical would have established the rates at which this seem- reviews on sertraline,14 it is clear that some of the sui- ingly new phenomenon might be happening. How- cides and suicidal acts categorized as occurring while ever, no such studies have ever been undertaken. This patients were taking placebo actually occurred during review, therefore, will in lieu cover the RCT data on a placebo washout period; placebo and washout sui- newly released antidepressants and suicidal acts, the cides are therefore distinguished here. Third, data for meta-analyses of efficacy studies in depression that citalopram, from another article by Khan et al,15 are in- have been brought to bear on the question and relevant cluded (although no details about the validity of as- signments to placebo are available). Fourth, fluoxetinedata from public domain documents are presented, Efficacy studies
again dividing the data into placebo and washoutperiod suicidal acts, along with data for venlafaxine.16 In lieu of specifically designed RCTs, the RCTs that When washout and placebo data are separated and formed the basis for the licence application for recent analyzed in terms of suicidal acts per patient (ex- antidepressants are one source of data. Khan and col- cluding missing bupropion data) using an exact Man- leagues10 recently analyzed RCT data to assess whether tel-Haenszel procedure with a 1-tailed test for signifi- it was ethical to continue using placebos in antidepres- cance, the odds ratio of a suicide while taking these sant trials. Although the US Food and Drug Adminis- new antidepressants as a group compared with pla- tration (FDA), in general, recommend that data from cebo is 4.40 (95% confidence interval [CI] 1.32-infinity; clinical trials be analyzed both in terms of absolute p = 0.0125). The odds ratio for a suicidal act while tak- numbers and patient exposure years (PEY), given that ing these antidepressants compared with placebo is an assessment of the hazards posed by placebo was the 2.39 (95% CI 1.66-infinity; p ≤ 0.0001). The odds ratio object of this study, the investigators appropriately for a completed suicide while taking an SSRI antide- Rev Psychiatr Neurosci 2003;28(5)
pressant (including venlafaxine) compared with mixture of trials. The current analysis limits the num- placebo is 2.46 (95% CI 0.71-infinity; p = 0.16), and the ber of studies but ensures that they are roughly com- odds ratio for a suicidal act while taking SSRIs com- parable, and the selection of studies is based on regula- pared with placebo is 2.22 (95% CI 1.47-infinity; p ≤ tory requirements rather than individual bias.
If washout suicidal acts are included with placebo, as Meta- and other analyses of SSRIs
the companies appear to have done, but adjusting the and suicidal acts
denominator appropriately, the relative risk of suicidalacts while taking sertraline, paroxetine or fluoxetine In addition to the RCT data indicating an excess of sui- compared with placebo becomes significant, with fig- cidal acts by those taking SSRIs, the clinical trials on ures ranging from 3.0 for sertraline to over 10.0 for flu- zimelidine, the first SSRI, suggested there were more suicide attempts by patients taking it than by those Other data sets yield similar findings. For instance, in taking comparators, but Montgomery and colleagues18 Pierre Fabre's clinical trial database of approximately reported that although this might be the case, zimeli- 8000 patients, the rate for suicidal acts by those taking dine appeared to do better than comparators in reduc- SSRIs appears to be 3 times the rate for other antide- ing already existing suicidal thoughts. A similar analy- pressants.17 However, these other data sets include a sis demonstrated lower suicide attempt rates for thosetaking fluvoxamine than the comparators in clinicaltrials.19 Problems with paroxetine led to similar analy- Table 1: Incidence of suicides and suicide attempts in
antidepressant trials from Khan et al10,15 and Kirsch et al16
The best-known analysis of this type was published by Eli Lilly after the controversy with fluoxetine emerged; from the analysis of pooled data from 17 double-blind clinical trials in patients with major depressive disorder, the authors concluded that "data from these trials do not show that fluoxetine is asso- ciated with an increased risk of suicidal acts or emer- gence of substantial suicidal thoughts among depressed patients."22 There are a number of method- ological problems with Lilly's analysis, however, and these apply to some extent to all other such exercises.
First, none of the studies in the analysis were designed to test whether fluoxetine could be associated with the emergence of suicidality. In the case of fluoxetine, all of the studies had been conducted before concerns of sui- cide induction had arisen. Some of the studies used in the analysis had, in fact, been rejected by the FDA.
Second, only 3067 patients of the approximately 26 000 patients entered into clinical trials of fluoxetine were included in this meta-analysis. Third, no mention was made of the fact that benzodiazepines had been co- prescribed in the clinical trial program to minimize the agitation that Lilly recognized fluoxetine could cause.8 All investigational drugs
Fourth, no reference was made to the 5% of patients All SSRIs*
who dropped out because of anxiety and agitation.
Given that this was arguably the very problem that was at the heart of the issue, the handling of this issue was not reassuring. The 5% dropout rate for agitation *SSRI = selective serotonin reuptake inhibitor.
or akathisia holds true for other SSRIs as well, and the J Psychiatry Neurosci 2003;28(5)
differences between SSRIs and placebo are statistically suicidality. In a fourth study of 643 patients, conceived significant. Given that the Diagnostic and Statistical 20 years before fluoxetine was launched and instituted Manual of Mental Disorders, fourth edition, text revision 10 years before launch, only 185 patients received flu- (DSM-IV-TR) has connected akathisia with suicide risk, oxetine at any point.29 This was clearly not a study designed to establish whether fluoxetine might induce Finally, this and other analyses depend critically on suicidality. None of these studies fit the definition of item 3 (i.e., suicide) of the Hamilton Rating Scale for Depression; this approach to the problem is one that Although not properly epidemiological, 2 post- FDA officials, Lilly personnel and Lilly's consultants8 marketing surveillance studies that compared SSRI agreed was methodologically unsatisfactory. The argu- with non-SSRI antidepressants found a higher rate of ment in these meta-analyses has, broadly speaking, induction of suicidal ideation for those taking SSRIs, been that in the randomized trials, the SSRI reduced although not in the rates of suicidal acts or suicides.30,31 suicidality on item 3 and that there was no emergence In a more standard epidemiological study of 222 sui- of suicidality, as measured by this item. To claim that cides, Donovan et al32 reported that 41 of those suicides the prevention of or reduction of suicidality in some were committed by people who had been taking an patients in some way means that treatment cannot pro- antidepressant in the month before their suicide; there duce suicidality in others is a logical non sequitur. The was a statistically significant doubling of the relative argument that item 3 would pick up emergent suicidal- risk of suicide in those taking SSRIs compared with tri- ity in studies run by clinicians who are not aware of this cyclic antidepressants. In a further epidemiological possible adverse effect has no evidence to support it.
study of 2776 acts of deliberate self-harm, Donovan et Despite these methodological caveats, the claim that al33 found a doubling of the risk for deliberate self- SSRIs reduce suicidality in some patients appears harm for those taking SSRIs compared with other anti- strong. However, insofar as SSRIs reduce suicidal acts in some, if there is a net increase in suicidal acts asso- A set of post-marketing surveillance studies carried ciated with SSRI treatment in these same trials, the out in primary care in the United Kingdom by the Drug extent to which SSRIs cause problems for some pa- Safety Research Unit (DSRU)34 recorded 110 suicides in tients must be greater than is apparent from consider- over 50 000 patients being treated by general practition- ers in Britain. The DSRU methodology has since beenapplied to mirtazapine, where there have been 13 sui- Epidemiological studies
cides reported in a population of 13 554 patients.35 Thispermits the comparisons outlined in Table 2.
Epidemiology traditionally involves the study of repre- A further study from British primary care was un- sentative samples of the population and requires a dertaken by Jick and colleagues,36 who investigated the specification of the methods used to make the sample rate and means of suicide among people taking com- representative. A series of what have been termed epi- mon antidepressants. They reported 143 suicides demiological studies have been appealed to in this among 172 580 patients taking antidepressants and debate. The first is a 1-column letter involving no sui- found a statistically significant doubling of the relative cides.24 The second is a selective retrospective post-marketing chart review25 involving no suicides, which Table 2: Drug Safety Research Unit studies of selective
analyzed by the American College of Neuropsy- serotonin reuptake inhibitors (SSRIs) and mirtazapine in
chopharmacology, the FDA and others,26,27 shows a 3- primary care practice in the United Kingdom
fold increased relative risk of emergent suicidality for fluoxetine versus other antidepressants.
A third study was conducted by Warshaw and Keller28 on patients with anxiety disorder, in which the only suicide was committed by a patient taking fluoxe- tine. However, only 192 of the 654 patients in this Total SSRIs
study received fluoxetine. This, therefore, was not a study designed to test fluoxetine's capacity to induce Rev Psychiatr Neurosci 2003;28(5)
risk of suicide with fluoxetine compared with the refer- 100 000 patients in non-hospitalized depression. An- ence antidepressant, dothiepin, when calculated in other primary care study from the Netherlands gives a terms of patient exposure years. Controlling for con- suicide rate of 33 per 100 000 patient years.38 Finally, Si- founding factors such as age, sex and previous suicide mon and VonKorff39 in a study of suicide mortality attempts left the relative risk at 2.1 times greater for among individuals treated for depression in Puget fluoxetine than for dothiepin and greater than any Sound, Wash., reported 36 suicides in 62 159 patient other antidepressant studied, although statistical years. The suicide risk per 100 000 patient years was 64 significance was lost in the process. Of further note are among those who received outpatient specialty mental the elevated figures for mianserin and trazodone, health treatment, 43 among those treated with antide- which are closely related pharmacologically to mirtaza- pressant medications in primary care and 0 among pine and nefazodone. Controlling for confounding those treated in primary care without antidepressants.
factors in the case of mianserin and trazodone, how- Utilizing a database of 2.5 million person years and ever, led to a reduction in the relative risk of these 212 suicides from North Staffordshire, Boardman and Healy40 modelled the rate for suicide in treated or un- To provide comparability with other figures, I have treated depression and found it to be of the order of recalculated these data in terms of absolute numbers 68/100 000 patient years for all affective disorders.40 and separated the data for fluoxetine (Table 3). The This rate gives an upper limit on the suicide rate in data in the Jick study, however, only allow compar- mood disorders that is compatible with observed isons between antidepressants.36 They shed no light on national rates of suicide in the United Kingdom. Board- the differences between treatment with antidepressants man and Healy estimate a rate of 27 suicides per and non-treatment or on the efficacy of antidepressants 100 000 patients per annum for primary care primary in reducing suicide risk in primary care. The traditional affective disorders. Possible relative risks for SSRIs figures with which the DSRU studies and the Jick from the DSRU studies set against these figures and study might be compared are a 15% lifetime risk for the findings from the Jick study for all antidepressants suicide for affective disorders. This would be inappro- excluding fluoxetine are presented in Table 4.
priate, however, because this 15% figure was derived Comparing the figures for SSRIs from Table 2 with from patients with melancholic depression in hospital those for the non-SSRI antidepressants from the Jick study gives a mean figure for non-SSRI antidepressants There are very few empirical figures available for of 68 suicides per 100 000 patients exposed compared suicide rates in primary care depression, the sample with a figure of 212 suicides for the SSRI group. Based from which the Jick et al36 and DSRU34 data come. One on an analysis of 249 803 exposures to antidepressants, study from Sweden37 reports a suicide rate of 0 per therefore, the broad relative risk on SSRI antidepres-sants compared with non-SSRI antidepressants or evennon-treatment is 234/68 or 3.44.
Table 3: Suicides rates of patients taking antidepressants in
primary care settings in the United Kingdom*
There are 2 points of note. First, these low rates for suicide in untreated primary care mood disorder pop- Table 4: Relative risk (RR) of suicide while taking SSRIs (from
DSRU studies) compared with general risk of suicide in UK
primary care primary affective disorders40 and in UK primary
care depression treated with non-SSRI antidepressants36
Total SSRI
Note: CI = confidence interval.
*From Jick et al.36 Note: DSRU = Drug Safety Research Unit.
J Psychiatry Neurosci 2003;28(5)
ulations are consistent with the rate of 0 suicides in in children and adolescents during fluoxetine treatment. J Am those taking placebo in antidepressant RCTs. Second, Acad Child Adolesc Psychiatry 1991;30:171-6.
3. Masand P, Gupta S, Dwan M. Suicidal ideation related to flu- correcting the DSRU figures for exposure lengths gives oxetine treatment [letter]. N Engl J Med 1991;324:420.
figures for suicides on sertraline and paroxetine com- 4. Rothschild AJ, Locke CA. Re-exposure to fluoxetine after seri- parable to those reported from RCTs by Khan et al.10 ous suicide attempts by 3 patients: the role of akathisia. J ClinPsychiatry 1991;52:491-3.
5. Creaney W, Murray I, Healy D. Antidepressant induced suici- Conclusion
dal ideation. Hum Psychopharmacol 1991;6:329-32.
6. Wirshing WC, Van Putten T, Rosenberg J, Marder S, Ames D, Hicks-Gray T. Fluoxetine, akathisia and suicidality: is there a Since antidepressant drug treatments were intro- causal connection? Arch Gen Psychiatry 1992;49:580-1.
duced, there have been concerns that their use may 7. Lane RM. SSRI-induced extrapyramidal side effects and lead to suicide.41 Hitherto, there has been a legitimate akathisia: implications for treatment. J Psychopharmacology public health concern that the debate about possible 8. Healy D. A failure to warn [editorial]. Int J Risk Safety Med hazards might deter people at risk from suicide from seeking treatment, possibly leading to an increased 9. Committee on Safety of Medicines. Current problems (no. 21).
number of suicides. The data reviewed here, however, Guidance on benzodiazepines. London (UK): Committee onSafety of Medicines and the Medicines and Healthcare Prod- suggest that warnings and monitoring are more likely to reduce overall risks or that at least we should adopt 10. Khan A, Warner HA, Brown WA. Symptom reduction and a position of clinical equipoise on this issue and re- suicide risk in patients treated with placebo in antidepressantclinical trials. Arch Gen Psychiatry 2000;57:311-7.
solve it by means of further study rather than on the 11. Laughren TP. The scientific and ethical basis for placebo- controlled trials in depression and schizophrenia: an FDA per- The evidence that antidepressants may reduce suicide spective. Eur Psychiatry 2001;16:418-23.
risk is strong from both clinical practice and RCTs. An 12. Storosum JG, van Zwieten BJ, van den Brink W, Gersons BP, Broekman AW. Suicide risk in placebo-controlled studies of optimal suicide reduction strategy would probably in- major depression. Am J Psychiatry 2001;158:1271-5.
volve the monitored treatment of all patients and some 13. Brecher M. FDA Review and Evaluation of Clinical Data Orig- restriction of treatment for those most at risk of suicide.
inal NDA 20-021, Paroxetine Safety Review. 1991 Jun 19.
14. Lee H. Statistical reviews on sertraline for FDA. 1990 Aug 14 In addition, given evidence that particular personality types suit particular selective agents and that mismatch- 15. Khan A, Khan SR, Leventhal RM, Brown WA. Symptom ing patients and treatments can cause problems,42 fur- reduction and suicide risk in patients treated with placebo inantidepressant clinical trials: a replication analysis of the Food ther exploration of this area would seem called for.
and Drug Administration database. Int J Neuropsychopharmacol2001;4:113-8.
Kirsch I, Moore TJ, Scoboria A, Nicholls SS. The emperor's new Competing interests: In recent years, Dr. Healy has had consultancies
drugs: an analysis of antidepressant medication data submitted to the with, been a principal investigator or clinical trialist for, been a chair- US Food and Drug Administration. Prevention and Treatment man or speaker at international symposia for or been in receipt of 2002;5:Article 23. Posted 15 Jul 2002. Available: www.
support to attend meetings from: Astra-Zeneca, Boots/Knoll Pharma- journals.apa.org/prevention/volume5/pre0050023a.html (ac- ceuticals, Eli Lilly, Janssen-Cilag, Lorex-Synthelabo, Lundbeck, Organon, Pharmacia & Upjohn, Pierre-Fabre, Pfizer, Rhone-Poulenc 17. Kasper S. The place of milnacipran in the treatment of depres- Rorer, Roche, SmithKline Beecham and Solvay. In the past 2 years, he sion. Hum Psychopharmacol 1997;12:S135-41.
has had lecture fees and support to attend meetings from Astra- 18. Montgomery SA, McAulay R, Rani SJ, Roy D, Montgomery Zeneca, and he has been an expert witness for the plaintiff in 4 legal DB. A double-blind comparison of zimelidine and amitripty- actions invoving SSRIs. He has also been consulted on a number of line in endogenous depression. Acta Psychiatr Scand 1981;63 other attempted suicide, suicide and suicide-homicide cases follow- ing antidepressant treatment, in the majority of which he has offered 19. Wakelin JS. The role of serotonin in depression and suicide.
the view that the treatment was not involved. He has also been an Do serotonin reuptake inhibitors provide a key? In: Gastpar expert witness for the National Health Service (NHS) in a series of M, Wakelin JS, editors. Selective serotonin reuptake inhibitors: novel or commonplace agents. Basel: Karger; 1988. p. 70-83.
20. Lopez-Ibor JJ. Reduced suicidality on paroxetine. Eur Psychia- References
21. Montgomery SA, Dunner DL, Dunbar G. Reduction of suicidal thoughts with paroxetine in comparison to reference antide- 1. Teicher MH, Glod C, Cole JO. Emergence of intense suicidal pressants and placebo. Eur Neuropsychopharmacol 1995;5:5-13.
preoccupation during fluoxetine treatment. Am J Psychiatry 22. Beasley CM, Dornseif BE, Bosomworth JC, Sayler ME, Ram- pey AH, Heiligenstein JH, et al. Fluoxetine and suicide: a 2. King RA, Riddle MA, Chappell PB, Hardin MT, Anderson meta-analysis of controlled trials of treatment for depression.
GM, Lombroso P. Emergence of self-destructive phenomena Rev Psychiatr Neurosci 2003;28(5)
23. American Psychiatric Association. Diagnostic and statistical et al. Deliberate self-harm and antidepressant drugs. Investi- manual of mental disorders, fourth ed., text rev. Washington: gation of a possible link. Brit J Psychiatry 2000;177:551-6.
American Psychiatric Association; 2000. p. 801. 34. MacKay FJ, Dunn NR, Wilton LV, Pearce GL, Freemantle SN, 24. Ashleigh EA, Fesler FA. Fluoxetine and suicidal preoccupation Mann RD. A comparison of fluvoxamine, fluoxetine, sertraline [letter]. Am J Psychiatry 1992;149:1750.
and paroxetine examined by observational cohort studies.
25. Fava M, Rosenbaum JF. Suicidality and fluoxetine: is there a Pharmacoepidemiol Drug Saf 1997;6(Suppl 3):235-46.
relationship? J Clin Psychiatry 1991;52:108-11.
35. Biswas P, Wilton LV, Shakir SAW. Pharmacovigilance of mir- 26. American College of Neuropsychopharmacology. Suicidal tazapine: results of a prescription event monitoring study of behavior and psychotropic medication. Neuropsychopharmacol- 13,554 patients in England [presentation]. Royal College of Psychiatrists Annual Meeting; 2001 Jul 9-13; London.
27. Teicher MH, Glod CA, Cole JO. Antidepressant drugs and the 36. Jick S, Dean AD, Jick H. Antidepressants and suicide. BMJ emergence of suicidal tendencies. Drug Safety 1993;8:186-212.
28. Warshaw MG, Keller MB. The relationship between fluoxetine 37. Hagnell O, Lanke J, Rorsman B. Suicide rates in the Lundby use and suicidal behavior in 654 subjects with anxiety disor- study: mental illness as a risk factor for suicide. Neuropsychobi- ders. J Clin Psychiatry 1996;57:158-66.
29. Leon AC, Keller MB, Warshaw MG, Mueller TI, Solomon DA, 38. van Weel-Baumgarten E, van Den Bosch W, van Den Hoogen Coryell W. Prospective study of fluoxetine treatment and sui- H, Zitman FG. Ten-year follow-up of depression after diagno- cidal behavior in affectively ill subjects. Am J Psychiatry sis in general practice. Br J Gen Pract 1998;48:1643-6.
39. Simon GE, VonKorff M. Suicide mortality among patients 30. Fisher S, Bryant SG, Kent TA. Postmarketing surveillance by treated for depression in an insured population. Am J Epi- patient self-monitoring: trazodone versus fluoxetine. J Clin Psychopharmacol 1993;13:235-42.
40. Boardman AP, Healy D. Modeling suicide risk in primary care 31. Fisher S, Kent TA, Bryant SG. Postmarketing surveillance by primary affective disorders. Eur Psychiatry 2001;16:400-5.
patient self-monitoring: preliminary data for sertraline versus 41. Kielholz P, Battegay R. Behandlung depressiver Zustands- fluoxetine. J Clin Psychiatry 1995;56:288-96.
bilder, unter spezieller Berücksichtigung von Tofranil, einem 32. Donovan S, Kelleher MJ, Lambourn J, Foster R. The occurrence neuen Antidepressivum. Schweiz Med Wochenschr 1958;88:763-7.
of suicide following the prescription of antidepressant drugs.
42. Tranter R, Healy H, Cattell D, Healy D. Functional variations Arch Suicide Res 1999;5:181-92.
in agents differentially selective to monoaminergic systems.
33. Donovan S, Clayton A, Beeharry M, Jones S, Kirk C, Waters K, Canadian College of Neuropsychopharmacology
Collège canadien de neuropsychopharmacologie
W.G. Dewhurst Travel Awards
The CCNP is making available up to 6 travel awards for research trainees who would not otherwise beable to attend the Annual Meeting (to be held in 2004 in Kingston, Ont.). The awards are for the leastexpensive airfare available (to be approved by the CCNP Treasurer) plus $250. Research trainees (gradu-ate students, postdoctoral fellows or clinical residents) working in Canada or Canadian research traineesworking abroad are eligible to receive the bursaries. Travel bursaries will be awarded to those whosubmit the best abstracts.
Those who wish to apply should send a completed abstract form with a letter of support from their researchsupervisor to: Ms. Rachelle Anderson, 1E7.19, Department of Psychiatry, University of Alberta, 8440-112 St., Edmonton AB T6G 2B7; Deadline: April 2, 2004. Applicants will be notified by April 19 of the decision of the Committee.
J Psychiatry Neurosci 2003;28(5)

Source: http://www.maths.bangor.ac.uk/research/ftp/biomed/03_17.pdf