How do general practitioners in Thailand diagnose and treat patients presenting with anxiety and depression?


*Manote Lotrakul, MD, Department of Psychiatry, Faculty of Medicine Ramathibodi Hospital, Rama 6 Road, Bangkok 10400, Thailand. Email:


Aims:  To examine general practitioners’ (GPs’) diagnosis of a case vignette presenting both anxiety and depression symptoms, and to understand their treatment preferences for the case.

Methods:  A total of 1193 copies of a questionnaire were sent to doctors in primary care settings throughout Thailand. The questionnaire inquired about GPs’ demographic information and training background, as well as common psychiatric diagnoses and drug prescriptions to patients in their practise. A case vignette of a patient presenting both anxiety and depression symptoms was then given, and GPs were asked to describe their diagnosis and treatment preferences. For comparison, postal questionnaires of the same case vignette were also sent to 40 psychiatrists practising in general hospitals, asking their opinion about the diagnosis and treatment preferences.

Results:  A total of 434 questionnaires (36.4%) were returned. GPs reported that 37.7% of their patients suffered from anxiety disorders while 28.4% suffered from depressive disorders. For the patient in the case vignette, GPs made a diagnosis of anxiety disorders (53.5%) more often than depressive disorders (31.9%), whereas the psychiatrists at the general hospitals made a diagnosis of depressive disorders (54%) more often than anxiety disorders (9.1%). One-third of the GPs prescribed only anxiolytics, while 15.4% prescribed only antidepressants. The most commonly prescribed antidepressant by GPs was amitriptyline, which 93% of GPs used at a dosage below 50 mg/day. Only 5.8% of them prescribed fluoxetine as antidepressant. The most frequently prescribed anxiolytic drug was diazepam (65.4%). The most common combination of drugs prescribed was amitriptyline and diazepam (38.7%).

Conclusion:  Compared to psychiatrists, GPs were more likely to diagnose anxiety than depression in patients with the same set of symptoms. They also preferred to use amitriptyline to treat depression, and prescribed the drug at a low dose. GPs in Thailand should be encouraged to prescribe fluoxetine for treatment of depression because it is safer and more convenient to use than tricyclic antidepressants.

IN THAILAND MOST patients with psychiatric disorders are treated by general practitioners (GPs) because of the limited number of psychiatrists in the country. Moreover, people are reluctant to seek help from a psychiatrist because it may imply that they have a mental illness. The impact of this stigma is evident even in developed countries.1–3 Thus, GPs in Thailand play an important role in taking care of patients with mental heath problems.

Regarding psychiatric disorders in primary care, results from most studies show that depression and anxiety disorders are common psychiatric conditions.4,5 Patients with depression or anxiety, however, often complain more about their physical symptoms than about their mental health conditions – a presentation that can obscure psychiatric diagnosis. Moreover, depressed patients seen in primary care often present with somatic symptoms and anxiety rather than typical depressive symptoms, as seen in psychiatric clinics.6–8 These often lead to an underrecognition or a misdiagnosis of depressive and anxiety disorders in primary care. It is therefore important that GPs be able to detect and manage these conditions appropriately.

So far, little is known about how GPs in Thailand diagnose and treat patients with anxiety and depression. The purpose of the present study was to examine GPs’ diagnosis of a hypothetical patient with both anxiety and depressive symptoms, and to understand their treatment preference. GPs’ perspectives on the prevalence of these two common psychiatric illnesses and their patterns of psychotropic drug prescription were also investigated.


This study was approved by the Ethics Committee on Human Experimentation of the Faculty of Medicine, Ramathibodi Hospital, Mahidol University. A current list of community hospitals was obtained from the Bureau of Health Service System Development, Ministry of Public Health. Questionnaires were sent to the community hospitals on the list, which consisted of 143 hospitals with only one doctor and 525 hospitals with several doctors. Hospitals in the latter group received two copies each of the questionnaire; a total of 1193 copies of the questionnaire were sent out.

The postal questionnaires for the present study included demographic questions for the GPs such as gender, age, years of experience in general practise, and their specialty training. GPs were asked to estimate the percentage of common psychiatric diagnoses in their practise. We also requested the percentage of anti-anxiety and antidepressant drugs prescribed and common drugs prescribed in each group. In the last section, GPs were asked to make a diagnosis of the patient in the case vignette. We also asked them to specify the medications and the usual dose they would prescribe for the case. The vignette was as follows.

‘Mrs A is a 40-year-old housewife who made a complaint to you that she had suffered from headache and fatigue for approximately 1 month. At the same time she often had sensations of shortness of breath, palpitations, lack of appetite, sleep difficulties and loss of interest. She also complained that excessive thinking gave her a headache. What worried her most was her husband's flirting behavior because if he had a new wife, she and her sons would be threatened by a lack of financial support. The physical examination revealed that she was quite thin and had mild pale conjunctiva. Vital signs and other findings were within normal limits.’

We also sent this case vignette to 40 psychiatrists who had been practising in general hospitals, asking them to make a diagnosis of the case and what drug treatment they would prescribe.

Statistical analyses were performed using SPSS version 10 (SPSS, Chicago, IL, USA). Frequency distribution, cross-tabulation, and Pearson χ2 tests were used for analyzing categorical data. Continuous data were analyzed on t-test and analysis of variance (ANOVA). The significance level was set at P < 0.05.


Four hundred and forty-four copies of the questionnaires were returned. Ten incomplete copies of the questionnaires were excluded from analysis, resulting in a final response rate of 36.4%. Among respondents, 57% were male; 81% had taken no further specialty training. The mean number of years they had been working was 5.2 ± 5.1 years (median, 3 years). Thirty-three psychiatrists (82.5%) also returned the questionnaires. They had been practising for a mean of 9.0 ± 8.3 years with a range of 1–30 years.

According to the GPs, the three most common psychiatric diagnosis made in their practise were anxiety disorders (37.7 ± 22.2%), alcohol and drug use disorders (28.6 ± 20.2%) and depressive disorders (28.4 ± 20.5%). According to t-test and ANOVA there were no statistically significant differences in the frequency of the diagnoses by GPs according to different background or practise setting, such as GP gender, medical school graduated from, number of patients examined per day, or number of working years. Of the total psychiatric drugs prescribed by GPs, 41.1 ± 20.9% were anxiolytics and 30.2 ± 19.9% were antidepressants. The most frequently prescribed anxiolytics were diazepam (72.9%), lorazepam (12.0%) and chlorazepate (10.4%). Among antidepressants, the classical tricyclic antidepressant, amitriptyline, was the most frequently prescribed drug (90.0%), while fluoxetine, a selective serotonin re-uptake inhibitor (SSRI) was rarely used (4.4%).

For the case vignette, 53.5% of GPs made a provisional diagnosis of anxiety disorders and one-third made a diagnosis of depressive disorders. Among psychiatrists, however, 54% made a diagnosis of depressive disorders, while only 9.1% made a diagnosis of anxiety disorders (Fig. 1).

Figure 1.

Percentage of diagnoses for the case vignette by (▪) general practitioners and (□) psychiatrists.

Table 1 shows that, for the case vignette, 84.6% of GPs and 84.8% of psychiatrists prescribed anxiolytic drugs. But although 66.8% of GPs prescribed antidepressants, almost all psychiatrists (97%) used them.

Table 1.  Antidepressants and anxiolytics prescribed for the case vignette
 DiazepamChlorazepateLorazepamAlprazolamClonazepamNo anxiolyticsTotal (%)
GPs (n = 434)       
 No antidepressants23.
Psychiatrists (n = 33)       
 No antidepressants0.

The most commonly prescribed antidepressant by GPs was amitriptyline (59.9%), whereas fluoxetine was prescribed by only 5.8% of GPs. The most frequently prescribed anxiolytic drug was diazepam (65.4%). The most common combination of drugs prescribed was amitriptyline and diazepam (38.7%). In addition to psychotropic drugs, most GPs also prescribed symptomatic medications such as paracetamol, propranolol, multivitamin, non-steroidal anti-inflammatory drugs and cyproheptadine.

For anxiolytic drug prescription, 55% of GPs prescribed diazepam at 2–5 mg/day and 34.1% prescribed it at 6–10 mg/day. Fifty percent of GPs prescribed lorazepam at 1 mg/day and 75% prescribed chlorazepate at 10 mg/day.

GPs who diagnosed depressive disorders used amitriptyline at higher doses than GPs who diagnosed anxiety disorders (P < 0.01). Compared to psychiatrists’ prescription, however, the dose of amitriptyline was relatively low (29.45 ± 14.3 mg/day for GPs and 47.5 ± 16.3 mg/day for psychiatrists). Most GPs (93.2%) prescribed amitriptyline at a dosage below 50 mg/day. There was no significant difference between the dosage of fluoxetine and diazepam for each diagnosis (Table 2).

Table 2.  Mean dose of each prescribed drugs for the case vignette
 Anxiety disorders (mean ± SD)Depressive disorders (mean ± SD)Mixed anxiety-depression (mean ± SD)P
  • Statistically significant difference (P < 0.01).

  • NS, not significant.

 Amitriptyline21.91 ± 10.129.45 ± 14.321.25 ± 9.0<0.01
 Fluoxetine27.50 ± 15.023.18 ± 9.024.00 ± 8.9NS
 Diazepam6.19 ± 2.65.62 ± 2.25.96 ± 2.4NS
 Amitriptyline10 ± 047.5 ± 16.320 ± 00.01
 Fluoxetine10 ± 020 ± 020 ± 0
 Diazepam7 ± 2.62.8 ± 3.47 ± 4.6NS


According to the Thai government regulations, newly graduated doctors have to work in rural areas for 1–3 years before they can pursue further medical specialty training.9 These physicians make up a substantial portion of the GPs in the country. Most of the respondents in the present study worked in the northeast, followed by the north – a proportion that corresponds with the general distribution of GPs working in community hospitals in Thailand.10

The present study shows that common psychiatric problems in primary care are anxiety disorders and depressive disorders –a result that is in line with other studies.4,5 While in other countries depressive illness in primary care is more prevalent than anxiety disorder,4,5 GP respondents in Thailand perceived anxiety disorder to be more common than depressive disorder. Possible explanations for this finding are either that GPs were less aware of depression than anxiety or that the prevalence of depression in primary care in Thailand was truly lower than anxiety. A comparison of responses to the case vignette by GPs and psychiatrists demonstrates that the first hypothesis is more likely. That is, when a patient comes to a doctor's office with both anxiety and depressive symptoms, alongside physical complaints – a familiar scenario in primary care in Thailand – the majority of psychiatrists are more likely to make a diagnosis of depressive disorder, while GPs are more likely to make a diagnosis of anxiety disorder.

The under-diagnosis of depression might stem from the possibility that depressive patients in primary care settings have symptoms that are not as typical as depressive symptoms that GPs had learned in medical schools.11 In such cases, particularly among patients who have anxiety symptoms, benzodiazepines provide rapid symptom improvement in the initial treatment period.12 Moreover, compared to amitriptyline, benzodiazepines are easier to administer and have a higher patient acceptance rate.13 A study of GPs in Sweden showed that individual experiences from general practise and private life are more influential on GPs than academic education and professional literature.14 Physicians gathered repertoires of understanding and actions from their experiences, and used these to guide their professional actions.14 The high patient acceptance rate of benzodiazepines and its rapid symptom improvement may be what leads GPs in Thailand to over-diagnose anxiety disorders and over-prescribe anxiolytic drugs. Further research should be conducted to verify this hypothesis.

Regarding the treatment of patient in the case vignette, GPs' practise was somewhat pragmatic. They placed more emphasis on treating the symptoms rather than attempting to establish a correct diagnosis. While half of them used both antidepressants and anxiolytics, a number of them also prescribed symptomatic medications such as muscle relaxants, propranolol and cyproheptadine. A previous study of benzodiazepine use among GPs in northern Thailand demonstrated that nearly half of GPs also use benzodiazepine in depressive patients.15 This might be due to GPs' lack of time, knowledge and skills to establish an appropriate diagnosis, so they address the problems by prescribing any possible medication that alleviates patient complaints.

GPs who made a diagnosis of depressive disorder with regard to the case vignette prescribed a statistically significantly higher dose of amitriptyline than GPs who made a diagnosis of anxiety disorder. This difference, however, was not considered clinically significant because amitriptyline at a dose below 50 mg/day is much lower than the recommended dose for treatment of depressive disorder.16 Among the antidepressants prescribed, amitriptyline was the drug of choice for the majority of GPs, even though it has many unfavorable side-effects and is difficult to increase to therapeutic dose. Such unfavorable side-effects and therapeutic dosage adjustment issues could have been avoided if GPs had considered prescribing fluoxetine instead of amitriptyline.

Fluoxetine is an SSRI antidepressant that is well tolerated compared with tricyclic antidepressants, so compliance is better, and treatment dropout is less frequent. A major advantage of fluoxetine over tricyclic antidepressants is its lower level of toxicity, especially at overdose.17,18 Besides treating depressive disorder, SSRI can also alleviate anxiety symptoms and treat several anxiety disorders.13,19,20 There is a belief among some clinicians that antidepressants with stimulating properties, such as fluoxetine, may cause more anxiety to emerge as a side-effect in depressed patients who have anxiety or agitation. But when considering anxiety as a function of depression, the data suggest that any antidepressants are equally effective for treating both depression and anxiety symptoms.21 Recently a study from an Asian country has shown that fluoxetine can be used to treat patients with anxiety and/or depression safely and cost-effectively in primary care settings of low-income countries.22 Currently, almost all SSRI are available in Thailand, including fluoxetine, fluvoxamine, sertraline, paroxetine and escitalopram. Fluoxetine has been available in Thailand for approximately 15 years and there is now a generic version available at a very cheap price (0.05USD/capsule). Attempts should be made to encourage GPs in Thailand to switch drug treatment from amitriptyline to fluoxetine. In uncertain cases, instead of prescribing the combination of amitriptyline and diazepam, GPs should be encouraged to use fluoxetine and/or diazepam.

Several limitations of the present study should be noted. First, there was a relatively low response rate. Responding physicians may differ from non-responding physicians in their attitudes toward anxiety and depression management, as well as in their practise. It is possible that physicians who responded to the survey had a particular interest in mental health issues. As such, their responses might not reflect those of the larger population of GPs. Unfortunately we did not mark surveys with identifiers. Although the lack of identifiers may have yielded more frank responses, it precluded comparisons between respondents and non-respondents.

Another limitation was that the present study was an opinion survey. The results reflected self-reported practises of physicians. There was no information regarding the extent to which self-reporting reflected what was actually done in practise. One should therefore be cautious about extrapolating actual prescribing behavior from these results.

In summary the present study has shown that GPs are more likely to over-diagnose anxiety disorders while under-diagnosing depressive disorders. When treating depression, GPs have a tendency to prescribe amitriptyline, but at a dosage that is lower than the therapeutic level. GPs should be encouraged to switch antidepressant prescription from amitriptyline to fluoxetine because fluoxetine is safer and easier to use than tricyclic antidepressants.


This study was partly supported by Grant 47018 of the Faculty of Medicine, Ramathibodi Hospital, Mahidol University, Bangkok, Thailand.