Hong Kong Journal of Psychiatry (1999) 9 (1) 15-20

ORIGINAL ARTICLE

pdf Full Paper in PDF

INTRODUCTION

Controlled clinical trials have shown that low or moderate doses of antipsychotic drugs, i.e. 300 to 700 mg of CPZ or its equivalent, are adequate for most patients in the treatment of acute episodes of schizophrenia (Rifkin et al., 1991; Smith, 1994). In the majority of cases, high-dose antipsychotic medication appears to offer no advantage over the standard dose regime (Baldessarini et al, 1988; Thomson, 1994).

However, optimal antipsychotic dosage for long term treatment · of chronic schizophrenic patients is less clear (Johnson, 1985). Recent dose reduction studies in chronic schizophrenia have proven that more than half of the patients can be maintained on minimal doses (100-300 mg CPZeq) without noticeable deterioration in their clinical condition (Faraone et al., 1989; Smith, 1994; Gilbert et al.,1995).

However, most pharmaco-epidemiological surveys have also revealed that actual clinical practice fails to conform to recent psychopharmacological guidelines (Michel & Kolakowska, 1981; Clark & Holden, 1987). Such disparities in prescribing practice also exist in Hong Kong. A local survey found that the mean antipsychotic dose for acutely ill schizophrenic patients treated in a general hospital psychiatric unit was 589 mg CPZ equivalent daily (Chiu et al., 1992) while chronic schizophrenic patients in a large mental hospital received a mean dose of 1049 mg CPZeq of antipsychotic medication (Chiu et al. , 1991). Patients in a long-stay rehabilitation facility were taking a surprisingly high mean dose of 698 mg CPZeq (Ungvari et al., 1996). All three local surveys detected prescribing practices characterized by polypharmacy, prolonged administration of anticholinergic antiparkinsonian agents and use of multiple divided doses of antipsychotics, although such prescription patterns are discouraged by contemporary textbooks and review articles (e.g. Baldessarini et al. , 1988, Potkin et al., 1993; Kane, 1995).

There are a number of clinical benefits to using crosssectional surveys. First, while limited in accuracy and generalizability, such surveys are easy to carry out and indicate a trend of prescribing habits which may reveal discrepancies between standard recommendations and actual clinical practice in a particular setting. Second, repeated at regular intervals, cross-sectional surveys may reveal harmful prescription trends in a given setting and facilitate a closer scrutiny of psychopharmacological treatment habits of individual teams. Finally, they may serve as one of the global indicators of quality of care. Surveys which provide indicators of rational pharmacotherapy are particularly important in the rehabilitation of chronic patients where the emphasis is usually placed on the psychosocial aspects of rehabilitation. Rational pharmacotherapy enhances the likelihood of successful rehabilitation for the chronic patients by providing optimal, albeit not absolute, symptom control while reducing side effects, improving drug compliance, reducing drug expenditure and saving nursing time.

Over the past 2 years we have made continuous efforts, to bring our prescribing practice more in line with international standards. To this end, we have paid special attention to rationalizing and, whereever possible, reducing psychotropic drug prescription. These objectives were addressed by careful assessment of every patient and his/her medication by the multidisciplinary team in weekly ward rounds. In this paper, we report the results of these efforts to change prescribing patterns for chronic psychiatric patients treated in our long-stay rehabilitation unit.

METHOD

The Rehabilitation Unit, which is part of the Department of Psychiatry, Chinese University of Hong Kong, has 166 beds served by a multidisciplinary team. All the patients referred to rehabilitation came from one of the territory's two large psychiatric institutions and an acute in-patient psychiatric unit of a general teaching hospital. The only exclusion criteria for entering rehabilitation in this facility were the presence of severe medical conditions, suicidal intent, and/or frequent aggressive/antisocial behaviour.

The following data were collected from drug charts and medical notes of all 158 patients treated at our Rehabilitation Unit on a randomly chosen census day (20 February 1996): age, sex, diagnosis, age of onset of illness, age of first hospitalization and current doses of psychotropic medication including those prescribed on 'as needed' basis provided they were given at least three times during the week preceding the census day. Antipsychotic doses were converted to CPZ equivalent doses in the same fashion as in previous local surveys (Davis, 1976; Ortiz & Gershon, 1986). Diagnoses were made according to the Diagnostic Criteria for Research, ICD-10 (WHO, 1993) by repeated clinical interviews and perusal of all available previous notes. For data analysis, t test (t), Mann-Whitney U test (MW) and Spearman rank correlation coefficient (rs) were used. For statistical analysis, significance was determined at p < 0.05 level.

RESULTS

One hundred and fifty-eight patients (51 female; mean age : 47.0, s.d.=14.8. and 107 male; mean age : 47.1 s.d.=13.2) were surveyed on census day. The distribution of primary psychiatric diagnoses was as follows : 136 (86.1%; 96 male and 40 female) schizophrenia, 9 epilepsy, 8 mental retardation, 2 alcohol dependence, 2 chronic mood disorder and 1obsessive compulsive disorder.

Thirty-three (20.9% ) patients, 22 male and 11 female, did not take any psychotropic medication. This nonmedicated group was significantly older than the rest of the cohort (mean : 55.9 us 44.7 years, t=3.26, p < .0001) and had a significantly later onset of illness (mean : 34, 1 vs 22.8 years, t=4.34, p< .0001). However, the two groups of patients did not differ in terms of the length of illness ( 21.6 vs 22.0 years, NS). All but one of the 15 patients in the non-medicated group with the diagnosis of schizophrenia were over the age of 60. Patients not having psychotropic medication were excluded from further analysis.

The remaining 125 (79 .1% ) patients received antipsychotic medication in a median dose of 300.0 mg CPZeq (range : 25-1500 mg) ; 30 (24% ) of them were on lowpotency and 95 (76%) m high-potency antipsychotics. Sixty (48.0% ) patients, all of them with the diagnosis of schizophrenia, were maintained on depot antipsychotics. In this group, 19 (31.7% ) were on depot alone, 34 (56.6% ) had one and 7 (11.6% ) had two additional oral antipsychotic drugs prescribed.

In terms of the number of antipsychotic medications, 71 patients (57% ) were on one, 47 (37.6%) were on two, and 7 (5.4% ) were on three antipsychotic agents at the same time. Of the 106 patients on oral medication, 51 (48.1%), 43 (40.6% ), 11 (10.3% ) and 1 (1% ) took antipsychotics once, twice, thrice and four times daily respectively. 53 (42.4% ) patients received an anticholinergic antiparkinsonian agent (benzhexol was the only available oral drug used) in a mean daily dose of 3.9 mg (range : 2-12 mg). Forty-six patients (35.3% ) were given psychotropic drugs other than benzhexol for a limited time in order to augment antipsychotic medication when it appeared to be ineffective. Such adjunctive psychotropics included benzo--diazepines [32 (25.6%) patients], antiepileptic [4 (3.2%) patients] and antidepressant [8 (6.4%) patients] drugs. Patients receiving adjunctive psychotropic drugs were on slightly lower doses of antipsychotics than those who were maintained on antipsychotics alone · (median : 250 mg us 400 mg CPZ equivalent; MW=l.20; NS) while there was no siginificant difference between the two groups with respect to benzhexol dosage (median : 0.0 mg us 0.0 mg; mean : 1.60 mg vs 1.68 mg; MW= 0.55; NS ).

Male patients received somewhat higher doses of antipsychotics (median : 350 mg us 200 mg CPZeq; MW= l.18. NS) and benzhexol (median : 0.0 mg vs 0.0 mg; mean : 1.68 mg vs 1.50 mg; MW=0.95. NS ) than females. Regarding the route of administration, patients on depot antipsychotics received higher doses than those on oral anti-psychotics (median : 420 mg vs 260 mg CPZ equivalent; MW= 0.82; NS) although the difference was not statistically significant. They were also given higher doses of benzhexol (median: 1.50 mg vs 0.00 mg; MW=2.45; p<.01). There was a statistically significant inverse correlation between the age and antipsychotic dosage (rs= -0.3388, p <.0001). Patients with a diagnosis of schizophrenia were administered higher antipsychotic (median : 350 mg vs O mg CPZ equivalent; MW=6.11 ; p<0.0001) and benzhexol (median : 0.0 mg vs 0.0 mg; mean : 1.52 mg vs 0.00 mg; MW= 3.54; p<.00005) doses compared with the nonschizophrenic group.

DISCUSSION

Antipsychotic treatment plays a pivotal role in the rehabilitation of chronic psychiatric patients. Therefore, periodical reviews of prescribing practice are essential to save the patients from unnecessary, and potentially harmful, medications. To maximize their potential for rehabilltation, it is very important to ensure that chronic psychiatric patients receive optimal drug treatment. Over the past two years we paid special attention to psychotropic drug prescription in our rehabilitation unit to enhance and maintain the staff's awareness of rational psychopharma-cotherapy. In addition to regular reviews of prescriptions carried out in the weekly multi-disciplinary ward rounds, seminars and lectures were offered and relevant literature was regularly distributed to all mental health professionals. Drug trials aimed at optimizing of maintenance treatment of Chinese patients have recently started.

These concerted efforts have yielded results in three aspects of our prescription practice. First, we succeeded in maintaining our patients on a reasonably low (median : 300 mg; mean : 341 mg CPZeq) daily antipsychotic dose. This dose falls within the recommended range (Baldessarini et al., 1988; Potkin et al., 1993; Kane, 1995) and is also similar to what has been reported for other patients in a rehabilitation facility (Morgan & Gopalaswamy, 1984). Potentially dangerous, aggressive or actively suicidal patients cannot be managed in our open-dorr rehabilitation setting. This fact partly explains the marked discrepancy in antipsychotic dosage between our unit and another local setting where chronic schizophrenic patients of similar age and illness duration received a mean dose of 1049 mg CPZ equivalent of antipsychotic medication (Chiu et al., 1991). In addition, 14% of our cohort, although chronically psychotic, suffered from psychiatric disorders other than schizophrenia. Our results should be interpreted in the light of a growing consensus that Asian schizophrenic patients require less antipsychotic medication than their Caucasian counterparts (Lin & Finder, 1983; Rosenblatt & Tang, 1987), a finding which may be accounted for by various pharmacokinetic and pharmacodynamic mechanisms (Lin et al., 1995). A recent study from China (Jin, 1994) reported that a 240 mg CPZeq mean daily dose sufficed for chronic schizophrenic patients participating in an active rehabilitation programme.

Second, as a consequence of low antipsychotic dosage, only 42% of our patients required benzhexol as opposed to 90% of chronic schizophrenic patients reported in an earlier local survey (Chiu et al., 1991). The role of anticholinergic antiparkinsonian agents in the long-term treatment of schizophrenia has not yet been clarified (Lavin & Rifkin, 1991). Recent guidelines drawn up by a panel of experts recommend their use only in the presence of parkinsonian symptoms and their gradual withdrawal after 3 months if the drug-induced Parkinson syndrome has subsided (WHO, 1990). The small dose of benzhexol our patients received reduces the likelihood of anticholinergic side effects. In a previous study (Chiu et a /.,1996; Ungvari et al., 1996) we found that less than 20% of our patients on benzhexol had manifestations of Parkinson syndrome indicating that even more patients should be withdrawn from beznhexol. Our results does not support the contention that Chinese patients are more vulnerable to antipsychotics-induced extrapyramidal side effects (Binder & Levy, 1981).

Third, we managed to maintain 33 (20.9% ) patients, among them 18 with schizophrenia, free of antipsychotic medication. The non-schizophrenic group consisted of 13 young epileptic and mentally retarded subjects, 1 patient with chronic affective psychosis and 1 patient with alcohol dependence. Not being permanently psychotic, some of these patients required only occasional sedation for which we preferred carbamazepine or valproic acid rather than antipsychotic drugs. Eighteen schizophrenic subjects (13% of all schizophrenic population surveyed) were kept free of antipsychotic medication. All of them were over 55 (range : 55-73 y) with no less than 20 years of illness duration each. These older schizophrenic patients had undergone various courses of pharmacological and ECT treatments in the past without achieving full symptomatic remission. In this late stage of their illness, continuous antipsychotic treatment did not significantly alter their condition. They have learned to live with their psychotic symptoms and shown adequate social adjustment. None of them had a history of significant violence and/or suicide attempts nor were they waiting for discharge back to the community. violence. We decreased their antipsychotic drugs very slowly over 2 years while carefully observing their mental status. We wish to emphasize that even gradual withdrawal of antipsychotic medication carries a 25-50% risk for relapse (Gilbert et al. 1995; Viguera et al, 1997) and can be cautiously tried only under special circumstances viz. advanced age, long duration of illness, stable symptomatology of mild severity, lack of history of violence and/or suicide attempts, reasonable adaptation to institutional life and permanent observation by mental health professionals.

The current study being a cross-sectional survey and not a controlled, standardized assessment, only a rough, global estimate can be given with respect to the patients' conditi0on following the rationalization of their drug regime. Compared to the previous 2-year period (1992-1994), there were more patients discharged (199/597 vs 51/378) and the number of suicides (2/597 vs 1/378), serious incidents (fights,indecent behaviour - 93/597 vs 77/378) and absconding patients (156/378 vs 156/597) did not change significantly during the study period (1994-1996). It is, therefore, safe to assume that the overall well-being of the patients as a group, if not improved, certainly did not get worse.

Despite our concerted efforts towards rational pharmacotherapy, some aspects of our current practice have remained a cause for concern. Depot neuroleptics were not only underutilized (only 48% of chronic patients received them) but they were also accompanied by one or two oral antipsychotic drugs in about two third of the cases. Such combination is not recommended except while switching from oral to depot medication, or temporarily adding oral medication to treat an acute exacerbation of psychosis (Glazer & Kane, 1992). Only 5 of our 60 patients on depot antipsychotics had a recent psychotic relapse, while for the remaining patients the concommittant use of oral medication seemed to be unjustified.

Despite regular medication reviews and subsequent correction of our prescription habits, a persistent pattern of polypharmacy was demonstrated with only 57% of our patients taking only one antipsychotic. However, further analysis is needed to determine if the concurrent use of more than one antipsychotic agent was probably justified in some rare cases. Jacob (1995) distinguished justified and unjustified polypharmacy in psychiatry. The justified use of multiple antipsychotics may occur temporarily during a period of overlap when changing from one drug to another. Also, use of several antipsychotic agents for treatmentresistent schizophrenic patients may be occasionally justified. Psychopharmacologists have until now strongly advocated using a single antipsychotic agent at any given time, claiming lack of therapeutic advantage with more than one agent and increased likelihood of side effects (Kane, 1995). With the introduction of atypical antipsychotics, such as clozapine and risperidone, which act on multiple neurotransmitter systems, the old dogma of strict monotherapy may need to be revised (Meltzer, 1992). Our strong clinical impression has been that a small number of chronic schizophrenic patients cannot be maintained on monotherapy, no matter how cautiously the second and third antipsychotic drug was withdrawn even while keeping the sum total of CPZeq daily dose constant. Patients as well as staff consistently reported an increase in psychotic symptoms and/or a noticeable deterioration in behaviour and work performance during such withdrawal. The reasons for the failure to rationalze pharmacotherapy in these cases are not clear - rigid, stereotyped behaviour, paranoid attitude are among the possibilities - and require a separate study. It is our conclusion that the effectiveness of polypharmacy for an individual treatment-resistent patient is more a matter of trial and error rather than reliance on scientifically sound and unequivocal guidelines (Meltzer, 1992; Jacob, 1995).

Adjunctive strategies targeting primarily the negative symptoms of treatment-resistent schizophrenic patients constitute a planned, and only transient, form of polypharmacy (Dufresne, 1995). These drug combinations are based on the assumption that the underlying neurobiological abnormalities in schizophrenia involve multiple neurotransmitter systems (Meltzer, 1992; Kane, 1995). Since atypical antipsychotics remain prohibitively expensive and their role in t.h.e long-term treatment of schizophrenia has not been clearly defined (Weiden et al , 1996), enhancing the effects of antipsychotic drugs with benzodiazepines, mood stabilizers or antidepressants is justified even though only a relatively small percentage of patients is expected to respond (Meltzer, 1992; Dufresne, 1995; Johns & Thompson, 1995). Again, given the lack of clear-cut guidelines, we feel that a time-limited trial with either of the above-mentioned drugs lasting usually 2-3 months should be offered to every treatment-resistent schizophrenic patient. It is an ethical choice provided the combination does not produce disturbing side effects. It should be noted that patients on adjunctive treatment required lower antipsychotic doses than those only on antipsychotic maintenance (median : 250 vs 400 mg CPZeq) although the difference did not reach statistically significant level.

Patients with the diagnosis of schizophrenia were given significantly higher doses of antipsychotics than nonschizophrenic subjects. This finding probably reflects the judicial use of antipsychotic medication in non-psychotic psychiatric disorders where antipsychotic agents were given only occasionally for sedation, mostly on a temporary basis. The inverse relationship between age and antipsychotic dosage indicates a justifiable caution in prescribing the elderly as most elderly patients with schizophrenia could function satisfactorily either on no drugs or on minimal doses. Male patients received slightly higher doses of antipsychotic (median : 380 vs 200 mg CPZeq) and antiparkinson (mean : 1.68 vs 1.50 mg benzhexol) drugs than female patients although the differences were not statistically significant. Staff perception of male patients as stronger, more aggressive and potentially more dangerous might account for the higher doses. Further studies controlling for variables such as age, weight and duration of illness are warranted to address the issue of gender-specific prescription habits.

The administration of antipsychotic drugs in divided daily doses reduces compliance and requires considerable nursing time for distribution. In maintenance treatment, most antipsychotics can be given once or twice a day (Kane, 1995). Our practice was, in fact, in line with this recommendation since only 12 (11% ) patients received their drugs 3 times or four times a day. Even in these patients the divided dose regime was used only temporarily while introducing a new antipsychotic drug.

Due to their cross-sectional design and the lack of correlation between drug prescription, the patients' medical and psychiatric condition and the characteristics of the particular setting, prescription pattern studies have their obvious limitations. Although results are difficult to generalize in the absence of an exact demographic and clinical description of the patient population in question, reports on prescription habits have mushroomed in the past 15 years since Michel & Kolakowska (1981) paper appeared. The popularity of such studies is understandable since these surveys form an the integral part of internal audit procedures. Since our surveys proved to be very useful by revealing important trends and pitfalls in our prescribing practices, we strongly recommend their implementation and regular use in other psychiatric facilities in Hong Kong

REFERENCES

Baldessarini, R.J., Katz, B. & Cotton, P. (1984). Dissimilar dosing with high-potency and low-potency neuroleptics. American Journal of Psychiatry, 141, 748-752.·

Baldessarini, R.J., Cohen, B,M. & Teicher, M.H. (1988).

Significance of neuroleptic dose and plasma level in the pharmacological treatment of psychoses. Archives of General Psychiatry, 45, 79-91.

Baldessarini, R. J., Kanda, J.C. & Centrorrino, F. (1995). Hospital use of antipsychotic agents in 1989 and 1993 : stable dosing with decreased length of stay. American Journal of Psychiatry, 152, 1038-1044.

Binder, R.L. & Levy, R. (1981). Extrapyramidal reactions in Asians. American Journal of Psychiatry, 138, 1243-1244.

Chiu, H.F.K., Shum, P.S. & Lam, C.W. (1991). Psychotropic drug prescribing to chronic schizophrenics in a Hong Kong hospital. International Journal of Social Psychiatry, 37, 187-191.

Chiu, H.F.K., Lee, S., Leung, C.M. & Wing, Y.K. (1992). Neuroleptic prescription for Chinese schizophrenics in Hong Kong. Australian and New Zealand Journal of Psychiatry, 26, 262-264.

Chiu, H.F.K., Ungvari, G.S., Pang, A.H.T. et al. A double-blind, placebo-controlled study on gradual withdrawal of antiparkinsonian medication in Chinese schizophrenic patients. Abstracts. Schizophrenia 1996. 4th International Congress, 6-9 October, 1996, Vancouver, Canada, p. 72.

Clark, A.F. & Holden, N.L. (1987). The persistence of prescribing habits: a survey and follow-up of prescribing to chronic hospital in-patients. British Journal of Psychiatry, 150, 88-91.

Davis, J.M. (1976). Comparative doses and costs of antipsychotic medication. Archives of General Psychiatry, 33, 858-861.

Dufresne, R. L. (1995). Issues in polypharmacotherapy : focus on depression in schizophrenia. Psychopharmacology Bulletin, 31, 789-796.

Faraone, S.V., Green, A.I., Brown, W., Yin, P. & Tsuang, M.T. (1989). Neuroleptic dose reduction in persistently psychotic patients. Hospital and Community Psychiatry, 40, 1193- 1195.

Gilbert, P.L., Harris, J., McAdams, L.A. & Jeste, D. V. (1995). Neuroleptic withdrawal in schizophrenic patients. Archives of General Psychiatry, 52, 173-182.

Glazer, W. M. & Kane, J.M. (1992). Depot neuroleptic therapy : an underutilized treatment option. Journal of Clinical Psy­ chiatry, 53, 426-433.

Jacob, K.S. (1995). The use of multiple psychotropic medication in the treatment of mental disorders. Australian and New Zea­ land Journal of Psychiatry, 29, 186-188.

Jin, Z. (1994). Effect of an open-door policy combined with a structured activity programme on the residual symptoms of schizophrenic in-patients. British Journal of Psychiatry, 165 (Suppl. 24), 52-57.

Johns, C.A. & Thompson, J.W. (1995). Adjunctive treatments in schizophrenia : pharmacotherapies and electroconvulsive therapy. Schizophrenia Bulletin, 21, 607-620.

Johnson, D.A.W. (1985) Antipsychotic medication : clinical guidelines for maintenance therapy. Journal of Clinical Psychiatry, 46, 6-15.

Kane, J. M. (1995). Clinical psychopharmacology of schizophrenia. In : Gubbard, G.O. ed. Treatment of Psychiatric Disorders. Washington: American Psychiatric Press.

Lavin, M. R. & Rifkin, A. (1991). Prophylactic antiparkinson drug use: II. Withdrawal after long-term maintenance therapy. Journal of Clinical Pharmacology, 31, 769-777.

Lin, K.M. & Finder, E. (1983). Neuroleptic dosage for Asians. American Journal of Psychiatry, 140, 490-491.

Lin, K.M., Anderson, D. & Poland, P.E. (1995). Ethnicity and psychopharmacology: bridging the gap. Psychiatric Clinics of North America, 18, 635-647.

Meltzer, H. Y. (1992) Treatment of the neuroleptic-nonresponsive schziophrenic patient. Schizophrenia Bulletin, 18, 515-542.

Michel, K. & Kolakowska, T. (1981) A survey of prescribing psychotropic drugs in two psychiatric hospitals. British Journal of Psychiatry, 138, 217-221.

Morgan, R. & Gopalaswamy, A.K. (1984). Psychotropic drugs: another survey of prescribing patterns. British Journal of Psychiatry, 144, 298-302.

Ortiz, A. & Gershon, S. (1986). The future of neuroleptic psychopharmacology. Journal of Clinical Psychiatry, 47, 3-11.

Potkin, S.G., Albers, L.J. & Richmond, G. (1993). Schizophrenia : an overview of pharmacological treatment. In : Dunner, D.L. ed. Current Psychiatric Therapy. Philadelphia : W.B. Saunders.

Reardon G.T., Rifkin, A., Schwartz, A (1989). Changing patterns of neuroleptic usage over a decade, American Journal of Psychiatry, 146, 726-729.

Rifkin, A., Doddi, S., Karajgi, B., Borenstein, M. & Wachspress, M. (1991). Dosage of haloperidol for schizophrenia. Archives of General Psychiatry, 48, 166- 170.

Rosenblatt, R. & Tang, S.W. (1987). Do Oriental psychiatric patients receive different dosages of psychotropic medication when compared with Occidentals. Canadian Journal of Psychiatry, 32, 270-274.

Smith, R. C. (1994). Lower-dose therapy with traditional neuroleptics in chronically hospitalized schizophrenic patients. Archives of General Psychiatry, 51, 427-429.

Thompson, C. (1994). The use of high-dose antipsychotic medication. British Journal of Psychiatry, 164, 448-458.

Van Putten, T. & Marder, S.R. (1987). Behavioral toxicity of antipsychotic drugs. Journal of Clinical Psychiatry, 48 (supp 9), 13-19.

Ungvari, G.S., Pang, A.HT., Chiu, H.F.K., Wong, C.K. & Lum, F.K. (1996). Psychotropic drug prescription in rehabilitation: a survey in Hong Kong. Social Psychiatry and Psychiatric Epidemiology, 31, 288-291.

Ungvari, G.S., Chiu, H.F.K., Lam, L.C.W. et al. Withdrawal of antiparkinsonian medication in chronic schizophrenic patients : a double-blind, placebo-controlled study. Abstracts of the Annual Scientific Meeting of the Hong Kong College of Psychiatrists, 15 December, 1996.

Viguera, A.C., Baldessarini, R.J., Hegarty, J.D., Van Kammen, D.P., Tohen, M. Clinical risk following abrupt and gradual with- drawal of maintenance neuroleptic treatment. Archives of General Psychiatry, 54, 49-55.

Weiden, P., Aquila, R. & Standard, J. (1996). Atypical antipsychotic drugs and long-term outcome in schizophrenia. Journal of Clinical Psychiatry, 57(Suppl 11), 53-60.

World Health Organization. (1990). Prophylactic use of anticholinergics in patients on long-term neuroleptic treatment. British Journal of Psychiatry, 156, 412.

World Health Organization (1993). ICD-10 Chapter V. Diag­ nostic Criteria for Research. Geneva : WHO.

L.Y. Chow MBChB MRCPsych Assistant Professor, Department of Psychiatry, Chinese University of Hong Kong.
Gabor S. Ungvari MD PhD FRANZCP FHKCPsych FHKAM{Psych) Associate Professor, Department of Psychiatry, Chinese University of Hong Kong.
Helen F. K. Chiu, MBBS, FRCPsych, FHKCPsych FHKAM(Psych) Associate Professor and Head, Department of Psychiatry, Chinese University of Hong Kong
Tony Leung, Msc Statistician, Department of Psychiatry, Chinese University of Hong Kong

*Correspondence : Dr LY. Chow, Department of Psychiatry, 11/F, Prince of Wales Hospital, Shatin, N. T. Hong Kong