Introduction

The past decade has seen an increase in the number and type of antidepressants available to psychiatrists and other clinicians. In Trinidad and Tobago there are at least 14 antidepressants available. It is important to better understand current prescribing practices and to what degree these practices reflect research findings.

Recent information suggests that psychiatrists chose antidepressants with more tolerable side effects₁ a decreased risk of harm in overdose, and better treatment efficacy₂. Numerous factors influence choosing an antidepressant drug therapy₃. The extent of influence of evidence-based medicine on physicians' practice and prescribing habits has been questioned ₄,₅ .

Less than a decade ago, studies on drug choice in Europe revealed that the tricyclic antidepressant (TCA) amitriptyline was the first-line treatment choice for most psychiatrists₆,₇. Huszonek and others described the trend of US retail pharmacy prescriptions by psychiatrists between 1986 and 1989, before and after the introduction of fluoxetine¹. In 1986, TCAs comprised 80% of the antidepressants prescribed, with amitriptyline and imipramine being the most often prescribed in 43.5% of the cases. This pattern changed by 1989, with TCAs comprising only 56% of retail prescriptions and fluoxetine accounting for 1 in 5 antidepressant prescriptions by psychiatrists¹. Olfson and others₈ surveyed US psychiatrists in 1993/1994 and found that the SSRIs were the preferred antidepressants in the outpatient setting in 63% of cases, and the TCA nortriptyline was used in only 7% of visits. The trend to use SSRIs more often than the older TCAs occurred without there being clear evidence to support greater efficacy or cost-effectiveness₉,₁₀,₁₁.

A previous study done in Trinidad by Moore and others₁₂ in1998/99 found that the TCAs, mainly amitryptiline, were the antidepressants most prescribed in outpatients psychiatric clinics in Trinidad and fluoxetine was the only selective serotonin reuptake inhibitor (SSRI) prescribed in these clinics.

The purpose of this study was to examine current prescribing practices of all practicing psychiatrists, psychiatric residents in postgraduate training and clinicians working in the psychiatric services for over five years in Trinidad and Tobago and to determine to what degree these practices reflect research findings.

Method

All practicing psychiatrists, postgraduate psychiatric residents in training, and doctors working in psychiatry for over five years in Trinidad and Tobago were asked to complete a self-administered questionnaire. A cross-sectional study design was employed and the study was conducted between September 1 2002 and November 30 2002. These doctors were chosen for the study, as they are responsible for treating the majority of psychiatric patients in government hospitals, public outpatient clinics and private clinics in Trinidad and Tobago.

The doctors were contacted by phone and the questionnaires were subsequently sent to them by mail, fax or hand delivered. Arrangements were made to have the completed questionnaires picked up. The survey included demographic data on age, gender and type of practice. It also covered first-line preferences in the treatment of depression and efficacy of antidepressants. Other variables were antidepressant agents most associated with the least and most side effects, antidepressant agents most associated with certain side effects including weight gain, sexual dysfunction, discontinuation syndrome, agitation and insomnia. The questionnaire also included items such as commonly used dosages of antidepressants, length of therapy for a first episode of depression, the factors that affect choice of antidepressant medication, combination of different classes of antidepressants, frequency of consultations and first-line preferences in the treatment of certain depressive subtypes. The questionnaire also included an item on the use of electroconvulsive therapy in the treatment of depression.

Results

Of the thirty-six questionnaires sent, there were 31 respondents, an overall response rate of 86%. Two respondents, both male psychiatrists, were excluded from the study. One indicated that he was in administration and had not prescribed medication for over five years and the other had not been actively practicing at the time of the study. The mean age of the respondents could not be calculated, as 52% (15/29) respondents did not state their age. Sixty-two % (18/29) were male and %(11/29) female. The respondents were 15 (52%) psychiatrists, 8 (28%) psychiatrists in training in the University of the West Indies D.M. postgraduate program in psychiatry and 6 (21%) doctors were neither psychiatrists nor psychiatrists in training, but were working in the psychiatric services for more than 5 years. The majority 55%(16/29) worked in a public outpatient and institutional setting, 42% (12/29) worked in a combined private-practice setting and a public outpatient and institutional setting, and 3% (1/29) was solely in private practice. The characteristics of nonresponders were not obtained.

The SSRIs were prescribed most frequently 15/29 (72%) in the treatment of a newly case of major depression followed by the SNRIs in 17% (5/29) (Table 1). Of the SSRIs, fluoxetine 52% (15/29) and paroxetine 24% (7/29) were favoured (Table 2). Eighty-six percent (25/29) indicated that the tricyclic that they used most frequently in the treatment of depression was amitryptiline (Table 3). Seven percent (2/29) indicated that they used the reversible MAOI, moclobemide, in the treatment of depression and 93% (27/29) did not use MAOIs (Table 4).

Figure 1

Table 1: Antidepressant group prescribed most frequently in the treatment of a new case of major depression

Figure 2

Table 2: Selective serotonin reuptake inhibitor (SSRI) prescribed most frequently in the treatment of major depression

Figure 3

Table 3: Tricyclic antidepressant (TCA) prescribed most frequently in the treatment of major depression

Figure 4

Table 4: Monoamine oxidase inhibitor (MAOIs) prescribed most frequently in the treatment of major depression

Fifty-five percent (16/29) indicated that the SSRIs were the antidepressants with greatest efficacy, 14% (4/29) indicated that the SNRIs were the antidepressants with greatest efficacy and 14% (4/29) stated that all antidepressants were equally efficacious (Table 5). Seventy-nine percent (23/29) indicated that the SSRI group of antidepressants had the least side effects (Table 6). Seventy-six percent (22/29) indicated that the tricyclic group of antidepressants had the most side effects (Table 7). Sixty-two percent (18/29) indicated that amitryptiline was the antidepressant most likely to cause weight gain (Table 8). Fluoxetine was endorsed as most likely to be associated with sexual dysfunction 55% (16/29), a discontinuation syndrome 35% (10/29), agitation 52% (15/29) and insomnia 55% (15/29).

Figure 5

Table 5: Antidepressant group with greatest efficacy

Figure 6

Table 6: Antidepressant group with least side effects

Figure 7

Table 7: Antidepressant group with most side effects

Figure 8

Table 8: Antidepressant most likely to cause weight gain

The most commonly prescribed daily dose of fluoxetine was 20 mg in 100% (29/29) (Table 13), of sertraline was 50 mg in 79% (23/29) (Table 14), of paroxetine was 20 mg in 86% (25/29) (Table 15), of citalopram was 20 mg in 55% (16/29) (Table 16), of venlafaxine XR was 75 mg in 86% (25/29) (Table 17), of amitryptiline was 50-100 mg in 65% (19/29) (Table 18), of imipramine was 50-100 mg in 45% (13/29) (Table 19) and tianeptine was 37.5 mg daily in 48% (14/29) (Table 20). Clomipramine was used by only 14% (4/29), nefazodone by 7% (2/29), maprotiline by 7% (2/29), moclobemide by 7% (2/29), reboxetine by 3% (1/25) and bupropion by 20% (6/29) (Tables 21, 22, 23, 24, 25)

Figure 9

Table 9: Antidepressant most likely to cause sexual dysfunction

Figure 10

Table 10: Antidepressant most likely to cause discontinuation syndrome

Figure 11

Table 11: Antidepressant most likely to cause agitation

Figure 12

Table 12: Antidepressant most likely to cause insomnia

Figure 13

Table 13: Most commonly prescribed daily dose fluoxetine

Figure 14

Table 14: Most commonly prescribed daily dose sertraline

Figure 15

Table 15: Most commonly prescribed daily dose paroxetine

Figure 16

Table 16: Most commonly prescribed daily dose citalopram

Figure 17

Table 17: Most commonly prescribed dose venlafaxine XR

Figure 18

Table 18: Most commonly prescribed dose amitryptiline

Figure 19

Table 19: Most commonly prescribed daily dose imipramine

Figure 20

Table 20: Most commonly prescribed daily dose tianeptine

Figure 21

Table 21: Most commonly prescribed daily dose clomipramine

Figure 22

Table 22: Most commonly prescribed daily dose nefazodone

Figure 23

Table 23: Most commonly prescribed daily dose maprotiline

Figure 24

Table 24: Most commonly prescribed daily dose moclobemide

Figure 25

Table 25: Most commonly prescribed daily dose bupropion

Fifty-two percent (15/29) clinicians stated that they would most likely prescribe an SSRI for major depression with fatigue (Table 26), 52% (15/29) also indicated that they would prescribe an SSRI for anxiety with depression (Table 27), 93% (27/29) indicated that they would prescribe an SSRI for panic disorder with depression (Table 28), 52% (15/29) indicated that they would prescribe SSRI for depression with suicidal ideation (Table 29), 24% (7/29) indicated that they would prescribe an SSRI for atypical depression (Table 30) and 59% (17/29) indicated that they would prescribe the TCA, amitryptiline for depression with prominent insomnia (Table 31).

Figure 26

Table 26: Antidepressants prescribed for major depression with fatigue

Figure 27

Table 27: Antidepressants prescribed for anxiety with depression

Figure 28

Table 28: Antidepressants prescribed for panic disorder with depression

Figure 29

Table 29: Antidepressants prescribed for depression with suicidal ideation

Figure 30

Table 30: Antidepressants prescribed for atypical depression

{image:31}

Eighty-three percent (24/29) treat an episode of depression for six months or more (Table 32). Previous response to treatment 97% (28/29), presence of medical illness 93% (27/29), side effects of drugs 93% (27/29), cost of antidepressants 90 % (26/29), age 86% (25/29), knowledge of drugs 79% (23/39) and availability of samples 76% (22/29) were all factors influencing the choice of antidepressant in the treatment of depression (Table 33). Eighty-three percent (24/29) respondents saw patients at least once every two weeks in the first month of treatment of an episode of depression (Table 35). Clinicians reported using between 1 to twelve antidepressants each.

{image:32}

{image:33}

{image:34}

{image:35}

Twenty-eight percent (8/29) combined antidepressants, with 24% (7/29) combining SSRI and TCA and 3% (1/29) combining SSRI and bupropion.

Thirty-eight percent (11/29) of respondents never used ECT and 24% (7/29) had not used ECT for more than five years. Only 10%(3/39) used ECT within the past year.

Discussion

The survey reflected the prescribing pattern for the period studied September 1 to November 30 2002. The data is similar to those of Petersen et al₁₃ and Olfson and others⁸ in both the choice of specific antidepressant agents and the preference for the class of SSRIs (72%) and SNRIs (17%) over the TCAs (7%). According to this study, SSRI agents are favoured as first line treatment in a first episode of depression and are prescribed more frequently than TCA agents. The SSRIs were probably the drug class of choice because the side effects are better tolerated than the older drugs, there is better patient compliance with the once a day dosing and the suicide potential is low. This finding is, however, inconsistent with those of a 2002 study by Moore and others¹² who found that TCAs, mainly amitryptiline, were the antidepressants most prescribed in outpatients psychiatric clinics in Trinidad and fluoxetine was the only selective serotonin reuptake inhibitor (SSRI) prescribed in these clinics. The discrepancy in these findings might be explained by the fact that the study by Moore and others¹² was done between 1998 and 1999 when fluoxetine was the only SSRI available in public outpatient clinics. This situation has since changed and the SSRIs, paroxetine and sertraline, were available at these clinics at the time this study was being done.

Although recent meta-analyses failed to find clinically important differences in efficacy between SSRIs and tricyclic antidepressants₁₄,₁₅,₁₆ and all antidepressants are believed to be equally efficacious for the treatment of depression₁₇, 55%(16/29) clinicians indicated a belief that SSRIs were most efficacious, 14% (4/29) indicated that the SNRI, venlafaxine was most efficacious and 14% (4/29) indicated that antidepressants were equally efficacious. Despite the lack of evidence of a significant difference in efficacy between older and newer agents, the majority of clinicians perceived the newer agents to be more efficacious than the older TCAs and monoamine oxidase inhibitors (MAOIs).

In general, there was a trend for respondents to prescribe the lowest recommended dosage of SSRIs, SNRIs, tricyclics and other antidepressants.

Respondents were using between 1 and 12 antidepressants each. Although the majority of respondents felt that the SSRIs were the most efficacious group of drug with the least side effects and 76% (22/29) indicated that the tricyclics caused most side effects, 93% (27/29) were currently prescribing amitryptiline. Additionally, although sexual dysfunction and agitation appear to occur at similar rates with all the SSRIs , fluoxetine was perceived to be most likely to cause these side effects.

Most respondents based their choice of antidepressant on factors such as previous response to treatment, presence of comorbid medical conditions, side effect profile, cost, age and availability of samples.

Despite inadequate evidence to support the efficacy of combining antidepressants and significant risks of drug-drug interactions leading to delayed metabolism and potentially dangerous serum levels (e.g. SSRIs + TCAs) ₁₈,₁₉, 24% (7/29) respondents indicated that they combined SSRIs and TCAs and 3% (1/29) combined SSRIs and bupropion.

The majority of respondents stated that they would prescribe TCAs for depression with prominent insomnia. For all other types of depression (major depression with fatigue, anxious depression, panic disorder with depression with suicidal ideation and atypical depression) SSRIs were deemed to be the class of drugs that would most likely be prescribed.

Most clinicians were treating a first episode of depression for an adequate period of six months or more and 38% (11/29) had never used electroconvulsive therapy in the treatment of depression.

The limitations of the study: 1. The opinions surveyed were subjective, and this limits attempts to understand this complex problem. 2. The response rate was not 100%.

Conclusions

At the time this study was conducted, the newer antidepressants, the SSRIs and the SNRIs were prescribed in the majority of cases and were considered to be of greater efficacy than the older antidepressants such as the TCAs and MAOIs. They were probably favored because their use was associated with fewer side effects and better patient compliance. Several factors, including previous response to treatment, influenced prescribing habits. Doctors in the psychiatric services were prescribing recommended dosages of antidepressants for an adequate time period.

These findings are significant as they highlight the discrepancy between empirical evidence and clinical practices and suggest that other factors influence clinicians' medication choices in the treatment of depression.