• dorsolateral prefrontal cortex;
  • major depressive disorder;
  • myoinositol;
  • N-acetyl aspartate;
  • proton magnetic resonance spectroscopy


  1. Top of page
  2. Abstract

Aim:  Recent neuroimaging studies support functional and structural alterations in the dorsolateral prefrontal cortex (DLPFC), particularly on the left side in patients with major depressive disorders (MDD). The aim of the present study was to examine the biochemical characteristics of left DLPFC as measured on proton (1H) magnetic resonance spectroscopy (MRS) in patients with drug-naïve first-episode MDD and a healthy control group. A second aim was to assess the effect of antidepressant treatment on the metabolites of DLPFC.

Methods:  Short-echo single-voxel 1H-MRS was done for the left DLPFC in 17 female drug-free MDD patients (mean age ± SD, 30.9 ± 6.9 years) and 13 matched control subjects (mean age ± SD, 29.1 ± 6.2 years) and was repeated at 8 weeks following antidepressant treatment.

Results:  Comparison of baseline values indicated that there were no significant differences in any of the metabolite ratios (N-acetyl aspartate/creatine [NAA/Cr], myoinositol [Ino]/Cr, and choline [Cho]/Cr) between patients and controls. Significant differences were detected between pre- and post-treatment Ino/Cr ratios (0.67 ± 0.13, 0.58 ± 0.22, P = 0.032, respectively), although there was no difference in NAA/Cr and Cho/Cr ratios.

Conclusion:  Although no significant metabolic alterations exist in female patients with drug-naïve first-episode MDD as evaluated on 1H-MRS, an increase in Ino/Cr was observed following 8-week antidepressant treatment. These findings give rise to the possibility that non-neuronal cells, particularly glial cells that are probably damaged, play a role in the action of antidepressant treatment.

MAJOR DEPRESSIVE DISORDER (MDD) is a severe, prevalent and debilitating mental disorder.1 The nature of the pathology that precipitates MDD, however, has yet to be identified. Recently, advances have been made in neuroimaging that have led to understanding of alterations in certain brain regions in patients with MDD.

Dorsolateral prefrontal cortex (DLPFC) is one of the brain structures that have been most commonly investigated in patients with MDD.2 The prefrontal cortex, a part of the limbic–thalamic–cortical network, participates in the regulation of mood, cognition and behavior.3 Neurobiological abnormalities in DLPFC, particularly on the left side, are believed to be involved in the pathophysiology of MDD.2 In patients with MDD, functional and structural neuroimaging studies indicate volumetric,4 functional,5,6 chemical7 and receptor binding8 alterations in DLPFC. Moreover, post-mortem studies report not only neuronal but also glial changes in this region.9 Despite these observations the nature of the biochemical abnormalities in DLPFC has not been clarified as yet.

Proton (1H) magnetic resonance spectroscopy (MRS) is a unique non-invasive neuroimaging tool that allows for the direct, in vivo and non-radioactive measurement of the metabolites in certain brain regions.101H-MRS studies of DLPFC in patients with MDD are limited and show inconsistencies. In the left DLPFC, N-acetyl aspartate (NAA) levels were found to be lower only in the late-onset11 and chronically ill12 patients. Choline/creatine (Cho/Cr) ratios were found to be higher in patients with MDD.11,13 Lower levels of Cho in pediatric patients, however, have been reported.14 Decreased levels of myoinositol (Ino) levels were determined in adult depressive subjects15,16 but not in the elderly depressed patients.11 Among them Ino, which is a marker of the glial cell,15 is most noteworthy. Growing evidence suggest that glial cell pathology is more marked in frontal cortex of MDD patients.17 Glial cells may also play a role in neurogenesis as well as neuronal support. Moreover, it is possible that the actions of antidepressants may be demonstrated via glial cells.18 To date no reports have been published exploring the metabolic changes in DLPFC following antidepressant medication in young adults.

The first aim of the present study was to examine the metabolite characteristics of DLPFC as measured on single-voxel, short-echo 1H-MRS at 3.0 T in patients with drug-naïve first-episode MDD and the healthy control group. The second aim was to determine the metabolite changes following antidepressant treatment.


  1. Top of page
  2. Abstract


Female outpatients with a diagnosis of first-episode MDD, ranging in age from 20 to 50 years, were recruited for the study. None of them had ever used any psychotropic drugs previously. Healthy women matched for age, weight, height, handedness and education were recruited via local advertisements. The study was approved by the local ethical committee. The study conformed to the code of ethics set out in the Declaration of Helsinki. Written informed consent was obtained from all subjects after the study was fully explained.

All subjects were assessed on a structured psychiatric interview (Structured Clinical Interview for Diagnosis for DSM-IV; SCID).19 Hamilton Depression Scale (HDRS) was used to determine the severity of the illness.20,21 All patients had scores of ≥17 on the 17-item HDRS.22 Items on the scale are scored 0–2 or 0–4. Scores of 0–7 are considered normal; 8–13, mild; 14–18, moderate; 19–22, severe; and ≥23, very severe.23

Exclusion criteria for all subjects included (i) presence of any DSM IV Axis I disorder (except MDD for the patients); (ii) history of use of psychotropic or corticosteroid medications; (iii) presence of any neurological disease, such as seizures, vascular disease or brain injury; (iv) presence of any physical illness (including hypertension, ischemic heart disease, diabetes, Cushing or obesity) as assessed on personal history, abnormal signs in clinical examination or laboratory data (complete blood count, biochemical tests, lipid profile, thyroidal function tests or electrocardiography screening); (v) pregnancy or lactation within 1 year; (vi) left-handedness as assessed on the Chapman and Chapman handedness inventory;24,25 and (vii) contraindications to magnetic resonance scans.


Seventeen female patients and 13 healthy females were involved in the study. An independent doctor, unconnected to this study decided on the patients’ medications and his instructions were followed. Patients and comparison subjects received comprehensive medical and neurological examinations. The medications that the patients were using after inclusion in the study and number of patients using that medication were as follows: 50 mg/day sertraline (n = 5), 10 mg/day escitalopram (n = 5), 20 mg/day paroxetine (n = 5), and 20 mg/day fluoxetine (n = 2). There were two dropouts in the patient group. They were on fluoxetine medication. Psychopathology and MR were performed at the baseline for all subjects, and depressed subjects were reassessed at 8 weeks of antidepressant treatment.

Imaging protocol

Sagittal T1-weighted spin-echo (relaxation time/echo time [TR/TE]; 650/20 ms), axial and coronal T2-weighted turbo spin-echo (TR/TE, 4000/100 ms) was done to exclude the presence of cranial abnormalities and to locate the voxel for MRS. Upon evaluation by an experienced neuroradiologist (KKO), 1H-MRS was performed. Single-voxel (1.5 × 1.5 × 1.5 cm3) 1H-MRS imaging (point resolved spectroscopy, TR/TE 3000/30 ms, no. averages: 192) was done with a 3.0-T scanner (Allegra, VA25, Siemens, Erlangen, Germany). The voxel was placed in the left DLPFC. The superior frontal sulcus, the lateral fissure, and the genu of the corpus callosum were used as anatomical boundaries for the voxel placement as described in previous studies (Fig. 1).12 After data acquisition, an automated program provided by the manufacturer was used for off-line post-processing on a workstation (Leonardo workstation, VD 10B; Siemens) and NAA/Cr, Cho/Cr, Ino/Cr ratios were calculated. 1H-MRS spectra are shown in Figure 2.


Figure 1. Magnetic resonance (MR) image showing location of MR spectroscopy voxel in the left dorsolateral prefrontal cortex.

Download figure to PowerPoint


Figure 2. Proton magnetic resonance spectroscopy spectra in the left dorsolateral prefrontal cortex at 3.0 T. Relaxation time/echo time: 3000/30 ms. Anatomical landmarks for volume of interest can be seen on the right-hand side. I, integral.

Download figure to PowerPoint

Statistical analysis

All analyses were performed using SPSS for Windows, version 11.0 (SPSS, Chicago, IL, USA). Two-tailed significance level was set at P < 0.05. Student's t-test was used to compare metabolite ratios between healthy subjects and patients. Metabolite ratios and clinical scores of patients before and after antidepressant treatment were compared with paired samples t-test. Pearson's correlation coefficient was used to correlate age and clinical variables within the measured metabolite levels.


  1. Top of page
  2. Abstract

Demographic characteristics of the patients and healthy subjects are presented in Table 1. Depressed patients and controls did not differ significantly in terms of age, education, height or weight. Mean duration of the illness was 4.4 ± 3.4 months. HDRS mean score at baseline was 23.2 ± 4.3 and this score is considered severe depression. This means that the patients are moderate–severely depressed.23 This scores was 11.8 ± 7.4 at 8 weeks of treatment. The HDRS scores decreased significantly (t = −8.47, d.f. = 16, P < 0.001) after treatment.

Table 1. Subject characteristics
 PatientsHealthy controlsStudent's t-test
Mean ± SDMean ± SD t P
  1. d.f.: 28; all patients and controls were female.

Age (years)30.9 ± 6.929.1 ± 6.20.740.463
Height (cm)163.2 ± 4.7166.3 ± 4.11.900.068
Weight (kg)60.6 ± 7.961.7 ± 4.30.450.653
Education (years)11.8 ± 2.613.4 ± 2.71.590.124

None of the metabolite ratios (NAA/Cr, Cho/Cr, and Ino/Cr) measured on 1H-MRS in the left DLPFC differed between patients and control subjects (Table 2). After the treatment, Ino/Cr ratios increased significantly (P = 0.032) compared to pretreatment values (Fig. 3). There was no significant difference, however, between pre- and post-treatment NAA/Cr and Cho/Cr ratios.

Table 2. 1H-MRS metabolite ratios for left DLPFC in drug-naïve female first-episode MDD patients
1H-MRS metabolite ratiosPatientsHealthy controls (n = 13)
Before treatment (n = 17)After treatment (n = 15)
  • TR/TE: 3000/30 ms.

  • Cho/Cr, choline/creatine; DLPFC, dorsolateral prefrontal cortex ; 1H-MRS, proton magnetic resonance spectroscopy; Ino/Cr, myoinositol/creatine; MDD, major depressive disorder; NAA/Cr, N-acetyl aspartate/creatine; TR/TE, relaxation time/echo time.

NAA/Cr1.86 ± 0.301.70 ± 0.241.81 ± 0.39
Cho/Cr0.78 ± 0.160.78 ± 0.190.84 ± 0.32
Ino/Cr0.58 ± 0.220.67 ± 0.130.69 ± 0.33

Figure 3. Myoinositol/creatine (Ino/Cr) ratio. Relaxation time/echo time, 3000/30 ms.

Download figure to PowerPoint

No correlation between metabolite ratios and HDRS scores before and after antidepressant treatment was observed. No other significant correlations were found between age, length of illness, education and any measured metabolite levels.


  1. Top of page
  2. Abstract

In the present study there was no significant difference in the major metabolite ratios (NAA/Cr, Cho/Cr, and Ino/Cr) of the left DLPFC between drug-naïve first-episode MDD patients and healthy controls on 1H-MRS. In contrast, after treatment a significant increase in Ino/Cr ratios was observed in the patients.

Prior 1H-MRS studies reported abnormalities in the left DLPFC of MDD patients, but they are few in number, and inconsistent. The lack of consensus among studies may be a consequence of variety of patient samples involved. Those investigations are not comparable in terms of age range (geriatric and pediatric groups), comorbid disease, medication status and lifetime drug and psychiatric history.11,13,14,26 Old and medically ill subjects are vulnerable to reactive gliosis and vascular incidents.27 Aging has been found to be associated with the decline in the level of metabolites in the brain.28 Information coming from post-mortem studies considered that evaluation of the brain metabolites of subjects aged ≤45 years might be more reliable.29,30 A strength of the present study concerned the sample group. First, the patient group had no comorbid axis I disorder, no lifetime history of psychiatric and medical disease or drug use. Second, only young adults were recruited. Last, the patient group was evaluated in their drug-naïve depressed state in their first episode. Furthermore, technical factors might have been another reason for inconsistent results of those studies. Most of the 1H-MRS data in the literature were acquired on 1.5-T MR systems, with limited spectral resolution from metabolites.31,32 Because metabolites are better resolved at higher field strength, and short echo time minimizes the signal loss due to transverse relaxation, in the present study MRS was done at 3.0 T field strength with short echo time. The metabolic findings of such a homogenous patient group were similar to those of the control subjects. These findings are similar to that of a recent study using 1H-MRS that found no difference in the metabolite levels (i.e. NAA, Cr, Cho) of the left DLPFC between young unmedicated unipolar disorder patients and healthy controls.12

It should be noted that a lower NAA/Cr value in the late-onset patients, who had more severe depression11 and an inverse correlation of NAA with duration of illness33 were reported in unipolar patients. The age and progression of illness might play a critical role in alterations of NAA. The present findings are in agreement with the results of the previous 1H-MRS studies of pediatric or drug-naïve young adult patients.12–14 Because NAA is predominantly a neuronal marker, these findings were in accordance with the observations of normal neuronal cell numbers in MDD.34

The Cho/Cr ratios did not differ between the patients and controls, and treatment had no significant effect on Cho/Cr upon antidepressant medication. Specifically in the left DLPFC, abnormally increased Cho levels in the elderly and treatment-naïve pediatric patients were observed,11,13 although in other studies no changes were observed.35 It was postulated by some of those authors that an increase in the Cho signals reflected increases in both membrane turnover and cellular signal transduction. The present findings of no change either in NAA/Cr or Cho/Cr may at least partly stem from the relative young age range (20–50 years) of the patients and the timing of imaging, which was done at the first episode.

Because Ino is mainly stored in glial cells, it is traditionally accepted as a marker of glial cells.15 Decreased Ino levels in DLPFC were found in adult depressive subjects,15,16 whereas some reports did not support those findings.11,14,35 Additionally, lower Ino levels in suicide victims were reported only in the frontal cortex but not in the occipital cortex or cerebellum.36 The role of glia in the etiology of MDD has not yet become clear. Lower post-mortem glial density has been reported in MDD.9,37 Despite the lack of differences at baseline in the present study, increases in the Ino/Cr ratios after treatment might also indicate the influence of antidepressant treatment on the probable damage in the glial cells. It should be kept in mind that monoamines released to the synapse by antidepressants could also be transported to the glial cells38. As we mentioned before, glial cells have their own trophic factors and play a role in signal regulation and neurogenesis.18 This topic needs further research.

The use of ratios for evaluation may be seen as a potential limitation, because changes could have resulted not only from NAA, Ino, Cho but also from the denominator (Cr). Cr level is considered to be relatively constant. A great number of studies supported unaltered Cr in MDD.11–14,33,39 Because of this it has been used as an internal standard for comparison.40 In the present study, if Cr had changed enough to affect the fraction, all metabolite ratios would have tended a similar way.

The present study has some limitations: Small sample size was the main limiting issue, as usually occurred in previous ones. In addition, although recruitment of only women into the groups ensures homogeneity, it restricts generalization of the findings. Also, use of a single-voxel method enabled a more accurate localization while limiting the findings in a restricted region. The authors are aware of inclusion, and thus the metabolic contribution of some adjacent white matter into the voxel, but the voxel size could not be decreased too much, to avoid decreases in the signal-to-noise ratios. Because there was no decision to control medication in the present study, inhomogeneous antidepressant treatment was another limitation. All drugs, however, belonged to the same group: selective serotonin re-uptake inhibitors.

In conclusion, the present study reports unaltered metabolite levels in the left DLPFC between drug-naïve women with first-episode MDD and healthy subjects. This is the first longitudinal study investigating effects of antidepressant medication using 1H-MRS in DLPFC of women patients with MDD. As a preliminary finding an increase in Ino/Cr ratios following treatment in the absence of any change in NAA and Cho demonstrate the role of glial cells in MDD. Future studies are needed in larger homogenous groups with a multi-voxel design to clarify the effect of medication on the metabolites in larger volumes of interest in the brain.


  1. Top of page
  2. Abstract

This project was supported by an award from the Psychiatric Association of Turkey.


  1. Top of page
  2. Abstract
  • 1
    Ustun TB, Ayuso-Mateos JL, Chatterji S, Mathers C, Murray CJ. Global burden of depressive disorders in the year 2000. Br. J. Psychiatry 2004; 184: 386392.
  • 2
    Davidson RJ, Pizzagalli D, Nitschke JB, Putnam K. Depression: Perspectives from affective neuroscience. Annu. Rev. Psychol. 2002; 53: 545574.
  • 3
    Fossati P, Ergis AM, Allilaire JF. Executive functioning in unipolar depression: A review. Encephale 2002; 28: 97107.
  • 4
    Hastings RS, Parsey RV, Oquendo MA, Arango V, Mann JJ. Volumetric analysis of the prefrontal cortex, amygdala, and hippocampus in major depression. Neuropsychopharmacology 2004; 29: 952959.
  • 5
    Fitzgerald PB, Oxley TJ, Leird AR, Kulkarni J, Egan GF, Daskalakis ZJ. An analysis of functional neuroimaging studies of dorsolateral prefrontal cortical activity in depression. Psychiatry Res. 2006; 148: 3345.
  • 6
    Narita H, Odawara T, Iseki E, Kosaka K, Hirayasu Y. Psychomotor retardation correlates with frontal hypoperfusion and the Modified Stroop Test in patients with major depression under 60 years old. Psychiatry Clin. Neurosci. 2004; 58: 389395.
  • 7
    Yildiz-Yesiloglu A, Ankerst DP. Review of 1H magnetic resonance spectroscopy findings in major depressive disorder: A meta-analysis. Psychiatry Res. 2006; 147: 125.
  • 8
    Bielau H, Steiner J, Mawrin C et al. Dysregulation of GABAergic neurotransmission in mood disorders: A postmortem study. Ann. N. Y. Acad. Sci. 2007; 1096: 157169.
  • 9
    Rajkowska G, Miguel-Hidalgo JJ, Wei J et al. Morphometric evidence for neuronal and glial prefrontal cell pathology in major depression. Biol. Psychiatry 1999; 45: 10851098.
  • 10
    Stanley JA. In vivo magnetic resonance spectroscopy and its application to neuropsychiatric disorders. Can. J. Psychiatry 2002; 47: 315326.
  • 11
    Kumar A, Thomas A, Lavretsky H et al. Frontal white matter biochemical abnormalities in late-life major depression detected with proton magnetic resonance spectroscopy. Am. J. Psychiatry 2002; 159: 630636.
  • 12
    Brambilla P, Stanley JA, Nicoletti MA et al. 1H Magnetic resonance spectroscopy study of dorsolateral prefrontal cortex in unipolar mood disorder patients. Psychiatry Res. 2005; 138: 131139.
  • 13
    Farchine TR, Moore GJ, Rosenberg DR. Proton magnetic resonance spectroscopic imaging in pediatric major depression. Biol. Psychiatry 2002; 52: 8692.
  • 14
    Caetano SC, Fonseca M, Olvera RL et al. Proton spectroscopy study of the left dorsolateral prefrontal cortex in pediatric depressed patients. Neurosci. Lett. 2005; 384: 321326.
  • 15
    Frey R, Metzler D, Fischer P, Heiden A. Myo-inositol in depressive and healthy subjects determined by frontal 1H-magnetic resonance spectroscopy at 15 Tesla. J. Psychiatr. Res. 1998; 32: 411420.
  • 16
    Gruber S, Frey R, Mlynarik V et al. Quantification of metabolic differences in the frontal brain of depressive patients and controls obtained by 1H-MRS at 3 Tesla. Invest. Radiol. 2003; 38: 403408.
  • 17
    Migual-Hidalgo JJ, Rajkowska G. Comparison of prefrontal cell pathology between depression and alcohol dependence. J. Psychiatr. Res. 2003; 37: 411420.
  • 18
    Pav M, Kovaru H, Fiserova A, Havrdova E, Lisa V. Neurobiological aspects of depressive disorder and antidepressant treatment: Role of glia. Physiol. Res. 2008; 57: 151164.
  • 19
    First RLS, Gibbon M, Williams JBW. Structured Clinical Interview of DSM IV Axis I Disorders SCID-I. Biometrics Research Department, New York State Psychiatric Institute, New York, 1997.
  • 20
    Hamilton M. Development of a rating scale; for primary depressive illness. Br. J. Soc. Clin. Psychol. 1967; 6: 278296.
    Direct Link:
  • 21
    Akdemir A, Turkcapar MH, Orsel SD, Demirergi N, Dag I, Ozbay MH. Reliability and validity of the Turkish version of the Hamilton depression rating scale. Compr. Psychiatry 2001; 42: 161165.
  • 22
    O’Leary D, Costello F, Gormley N, Webb M. Remission onset and relapse in depression, an 18-month prospective study of course for 100 first admission patients. J. Affect. Disord. 2000; 57: 159171.
  • 23
    Sadock B, Sadock V. Kaplan and Sadock's Comprehensive Textbook of Psychiatry, 7th edn, vol. 1 S. 769. Lippincott, Williams and Wilkins, Philadelphia, PA, 2000.
  • 24
    Chapman JP, Chapman LJ. Handedness of hypothetically psychosis-prone subjects. J. Abnorm. Psychol. 1987; 96: 8993.
  • 25
    Nalcaci E, Kalaycioglu C, Gunes E, Cicek M. Reliability and validity of a handedness questionnaire. Turk Psikiyatr. Derg. 2002; 13: 99106.
  • 26
    Murata T, Kimura H, Omori M et al. MRI white matter hyperintensities, 1H-MR spectroscopy and cognitive function in geriatric depression: A comparison of early- and late-onset cases. Int. J. Geriatr. Psychiatry 2001; 16: 11291135.
  • 27
    Davis S, Thomas A, Perry R, Oakley A, Kalaria RN, O’Brien JT. Glial fibrillary acidic protein in late life major depressive disorder: An immunocytochemical study. J. Neurol. Neurosurg. Psychiatry 2002; 73: 556560.
  • 28
    Fukuzako H, Hashiguchi T, Sakamoto Y et al. Metabolite changes with age measured by proton magnetic resonance spectroscopy in normal subjects. Psychiatry Clin. Neurosci. 1997; 51: 261263.
  • 29
    Miguel-Hidalgo JJ, Baucom C, Dilley G et al. Glial fibrillary acidic protein immunoreactivity in the prefrontal cortex distinguishes younger from older adults in major depressive disorder. Biol. Psychiatry 2000; 48: 861873.
  • 30
    Si X, Miguel-Hidalgo JJ, O'Dwyer G, Stockmeier CA, Rajkowska G. Age-dependent reductions in the level of glial fibrillary acidic protein in the prefrontal cortex in major depression. Neuropsychopharmacology 2004; 29: 20882096.
  • 31
    Moore GJ, Galloway MP. Magnetic resonance spectroscopy: Neurochemistry and treatment effects in affective disorders. Psychopharmacol. Bull. 2002; 36: 523.
  • 32
    Pfleiderer B, Michael N, Erfurth A et al. Effective electroconvulsive therapy reverses glutamate/glutamine deficit in the left anterior cingulum of unipolar depressed patients. Psychiatry Res. 2003; 122: 185192.
  • 33
    Michael N, Ertfuth A, Ohrmann P, Arolt V, Heindel W, Pfleiderer B. Metabolic changes within the left dorsolateral prefrontal cortex occurring with electroconvulsive therapy in patients with treatment resistant unipolar depression. Psychol. Med. 2003; 33: 12771284.
  • 34
    Ongur D, Drevets WC, Price JL. Glial reductions in the subgenual prefrontal cortex in mood disorders. Proc. Natl Acad. Sci. USA 1998; 95: 13 29013 295.
  • 35
    Binesh N, Kumar A, Hwang S, Mintz J, Thomas MA. Neurochemistry of late-life depression: A pilot two dimensional MR spectroscopic study. J. Magn. Reson. Imaging 2004; 20: 10391045.
  • 36
    Shimon H, Agam G, Belmaker RH, Hyde TM, Kleinman JE. Reduced frontal cortex inositol levels in post-mortem brain of suicide victims and patients with bipolar disorder. Am. J. Psychiatry 1997; 154: 11481150.
  • 37
    Rajkowska G. Postmortem studies in mood disorders indicate altered numbers of neurons and glial cells. Biol. Psychiatry 2000; 48: 766777.
  • 38
    Inazu M, Takeda H, Ikoshi H, Sugisawa M, Uchida Y, Matsumiya T. Pharmacological characterization and visualization of the glial serotonin transporter. Neurochem. Int. 2001; 39: 3949.
  • 39
    Michael N, Gosling M, Reutemann M et al. Metabolic changes after repetitive transcranial magnetic stimulation rTMS of the left prefrontal cortex: A sham-controlled proton magnetic resonance spectroscopy 1H MRS study of healthy brain. Eur. J. Neurosci. 2003; 17: 24622468.
  • 40
    Malhi SG, Valenzuela M, Wen W, Sachdev P. Magnetic resonance spectroscopy and its applications in psychiatry. Aust. N. Z. J. Psychiatry 2002; 36: 3143.