The new antidepressants have had a significant impact on prescribing practice as described in the article by John Tiller on page 92. Although fluoxetine was only marketed in the U.S.A. in 1988, by 1994 it was already the most commonly prescribed antidepressant.1 While the number of prescriptions of fluoxetine and the other 'new' antidepressants (moclobemide, paroxetine and sertraline) have yet to overtake the tricyclic antidepressants (TCAs) in Australia2, current trends would suggest that this is not unlikely in the near future.

Such a shift in prescribing significantly impinges upon government expenditure and, therefore, upon each individual taxpayer. In Australia, fluoxetine was listed on the Pharmaceutical Benefits Scheme in 1991, moclobemide in 1992, and paroxetine and sertraline in 1994. Between 1989 and 1993, the overall cost of antidepressants increased from $24 million to $64 million per year, while the total number of prescriptions remained unchanged.3 In the U.K., it has been estimated4 that if only 70% of TCAs were replaced by the selective serotonin reuptake inhibitors (SSRIs), the total annual cost of antidepressants in general practice would rise from £88 million to nearly £200 million.

Are the cost implications of this shift in prescribing practice justified? This question requires a review of the critical issues of comparative efficacy, tolerability and toxicity.

Comparative efficacy
There have now been 3 major meta analyses of the comparative efficacies of the SSRIs and TCAs.5,6,7 Each of these studies has reported no significant difference between the effectiveness of these antidepressant classes. For example, the largest analysis7 found that 50% of inpatients and 52% of outpatients responded to TCAs, while 54% of inpatients and 47% of outpatients responded to SSRIs - both non-significant differences. A similar meta analysis has found no differences between the effectiveness of the reversible MAOI moclobemide and the TCAs.8

These findings show that, for the majority of depressed patients, the new drugs are as effective as the old. In general practice, where most depression is of a mild to moderate severity, the new drugs would appear to be as effective as the TCAs.

However, there has been some discussion concerning relative efficacy in patients with melancholic or severe depression.1 Four major studies of melancholic inpatients found that the SSRIs9,10,11 and moclobemide12 are less effective than the TCAs. There are also studies which show no difference between the new antidepressants and the TCAs in severe depression.13 Therefore, uncertainty remains regarding the comparative effectiveness of the new drugs in patients with melancholic or severe depression.

One of the major advances of the new antidepressants is their lack of anticholinergic, cardiovascular and other adverse effects14, which were a major limitation of the TCAs.

However, do the different adverse effect profiles of the new drugs translate into improved tolerability? The most stringent test of this is surely the number of patients dropping out of treatment during a course of antidepressants. Reduced dropout rates have been claimed as one of the major economic advantages of these new medications.15

Three recent meta analyses now allow this issue to be addressed.6,16,17 First, regarding dropout rates due to adverse effects, the difference between the SSRIs and TCAs appears to be less dramatic than earlier literature suggested. Two of the three studies found significantly more dropouts due to adverse effects with TCAs than SSRIs, with figures of 19% vs. 15%16 and 19% vs. 14%17 - indicating approximately 25% less dropouts due to adverse effects with the newer agents. The third study6, however, only found a nonsignificant trend towards higher dropout rates in those on TCAs. Second, when total dropout rates due to any reason are examined, the differences are also less than would have been expected. One study6 found no difference between these antidepressant classes (TCAs 32% vs. SSRIs 33%). Another17found 10% fewer dropouts with SSRIs, with overall dropout rates of 33% with TCAs, compared to 31% with SSRIs - a significant, albeit small, difference.

The minimal toxicity of the new antidepressants (in terms of cardiac effects and lethality in overdose) is perhaps their most clear cut advantage. While deaths due to TCA overdose account for about 7% of all suicides4, deaths from SSRI overdose are so rare they warrant published case reports.18

Cost issues
The economics of evaluating pharmacological treatments for depression is a rapidly expanding science, albeit fraught with uncertainties regarding appropriate methodologies and the validity of particular assumptions.19 Recent attempts have been made to quantify the cost per life gained by using SSRIs to prevent suicides. Different models have been used to test various assumptions concerning rates of future suicide attempts, use of other suicide techniques and the proportion of total suicides prevented by SSRIs.4 The cost per life year gained through preventing suicides by the routine first line use of SSRIs is likely to be high, with estimates varying from £19 000 to £173 000, depending upon the assumptions used.

More recently, a report has been published examining suicide rates due to any cause (including violent methods such as hanging or shooting).20 Unexpectedly, the initial analysis indicated higher suicide rates on fluoxetine than other antidepressants available in the U.K. As this may have reflected a biased prescription of fluoxetine for those who were already suicidal or had failed to respond to previous antidepressants, the authors reanalysed the data after excluding such patients. This resulted in a finding of no difference in suicide rates on the various antidepressants. This study suggests that, while the new antidepressants are less toxic in overdose, they do not appear to reduce overall suicide rates.

Which antidepressant should I use?
How do these issues affect the general practitioner who decides to prescribe an antidepressant for a depressed patient?

On balance, current clinical evidence would suggest that, for the average mildly to moderately depressed patient seen in general practice, the new drugs are as effective as the old. The new drugs cause less severe adverse effects, although the difference is not as dramatic as pharmaceutical advertising would suggest. They are also much less likely to be lethal in overdose, but do not appear to reduce overall suicide rates. For the more severe melancholic patients, the advantages, if any, of the new drugs are not as clear cut.

Overseas trends would suggest that the prescribing of the new antidepressants will overtake the old. Although the patient pays no more (as these drugs are subsidised through the Pharmaceutical Benefits Scheme), we all indirectly pay the cost as taxpayers. The issues are complex. As clinicians, we must side with our individual patient's interests, but do the advantages of these new antidepressants justify the rising expenditure? The ongoing discussion in the next few years should be most intriguing.


Self-test questions

The following statements are either true or false.
Click anywhere on the panel for the answers.

1. The new antidepressants are more effective than tricyclic antidepressants in severe depression.

2. The new antidepressants reduce overall suicide rates in depressed patients significantly more than tricyclic antidepressants.



  1. Gram LF. Fluoxetine. N Engl J Med 1994;331:1354-61.
  2. Commonwealth Department of Human Services and Health. Australian statistics on medicines 1992. Canberra: Australian Government Publishing Service, 1994. \r\n
  3. Alchin TM, Tranby HW. Who bears the cost of antidepressants in Australia? Med J Aust 1994;161:547-9.
  4. Freemantle N, House A, Song F, Mason JM, Sheldon TA. Prescribing selective serotonin reuptake inhibitors as strategy for prevention of suicide. Br Med J 1994;309:249-53.
  5. Workman EA, Short DD. Atypical antidepressants versus imipramine in the treatment of major depression: a meta analysis. J Clin Psychiatry 1993;54:5-12.
  6. Song F, Freemantle N, Sheldon TA, House A, Watson P, Long A, et al. Selective serotonin reuptake inhibitors: meta analysis of efficacy and acceptability [see comments]. Br Med J 1993;306:683-7. Comments in: Br Med J 1993;306:1124-6; discussion 1126-7.
  7. US Department of Health and Human Services Depression Guideline Panel. Depression in primary care: Vol. 2: Treatment of major depression. Rockville, Maryland: US Department of Health and Human Services, 1993.
  8. Angst J, Stabl M. Efficacy of moclobemide in different patient groups: a meta analysis of studies. Psychopharmacology 1992;106(Suppl):S109-S113.
  9. Danish University Antidepressant Group. Citalopram: clinical effect profile in comparison with clomipramine. A controlled multi center study. Psychopharmacology 1986;90:131-8.
  10. Danish University Antidepressant Group. Paroxetine: a selective serotonin reuptake inhibitor showing better tolerance, but weaker antidepressant effect than clomipramine in a controlled multi center study. J Affect Disord 1990;18:289-99.
  11. Roose SP, Glassman AH, Attia E, Woodring S. Comparative efficacy of selective serotonin reuptake inhibitors and tricyclics in the treatment of melancholia. Am J Psychiatry 1994;151:1735-9.
  12. Danish University Antidepressant Group. Moclobemide: a reversible MAOA inhibitor showing weaker antidepressant effect than clomipramine in a controlled multi center study. J Affect Disord 1993;28:105-16.
  13. Nierenberg AA. The treatment of severe depression: is there an efficacy gap between SSRI and TCA antidepressant generations? J Clin Psychiatry 1994;55(Suppl A):55-59; discussion 601, 98-100.
  14. Mitchell PB. Selective serotonin reuptake inhibitors: adverse effects, toxicity and interactions. Adverse Drug React Toxicol Rev 1994;13:121-44.
  15. Le Pen C, Levy E, Ravily V, Beuzen JN, Meurgey F. The cost of treatment dropout in depression. A cost benefit analysis of fluoxetine vs. tricyclics. J Affect Disord 1994;31:1-18.
  16. Montgomery SA, Henry J, McDonald G, Dinan T, Lader M, Hindmarch I, et al. Selective serotonin reuptake inhibitors: meta analysis of discontinuation rates. Int Clin Psychopharmacol 1994;9:47-53.
  17. Anderson IM, Tomenson BM. Treatment discontinuation with selective serotonin reuptake inhibitors compared with tricyclic antidepressants: a meta analysis. Br Med J 1995;310:1433-8.
  18. Kincaid RL, McMullin MM, Crookham SB, Rieders F. Report of a fluoxetine fatality. J Anal Toxicol 1990;14:327-9.
  19. Burke MJ, Silkey B, Preskorn SH. Pharmacoeconomic considerations when evaluating treatment options for major depressive disorder. J Clin Psychiatry 1994;55(Suppl A):42-52; discussion 53-4, 98-100.
  20. Jick SS, Dean AD, Jick H. Antidepressants and suicide [see comments]. Br Med J 1995;310:215-8. Comment in: Br Med J 1995;310:205-6.

Philip B. Mitchell

Associate Professor, School of Psychiatry, University of New South Wales

Administrative Director, Mood Disorders Unit, Prince Henry Hospital, Sydney