Prescribing Second-Generation Antipsychotics and the Evolving Standard of Care in Italy

• Introduction
• Methods
• Study Area
• Study population
• Results
• Differences between patients treated with clozapine and patients treated with other second-generation APs
• Off-label prescribing of second-generation antipsychotics
• Discussion
• References

The present study was carried out to investigate the routine use of second-generation antipsychotic drugs in the Italian psychiatric care system. Seven outpatient psychiatric services enrolled a consecutive case series of patients who were being treated, or had started treatment, with clozapine, olanzapine, risperidone, or quetiapine. Information on sociodemographic and clinical variables, current psychotropic drug use, side-effects and past use of typical drugs was collected. In addition, patient symptoms and functional status were evaluated by the Health of the Nation Outcome Scale. Patients receiving off-label prescribing of second-generation antipsychotics were identified. A total of 209 patients were collected. In comparison with patients receiving other second-generation antipsychotics, living in residential facilities, unemployment, long psychiatric histories, and problems with activities of daily living and living conditions were more common in clozapine-treated patients. Nearly 80 % of patients receiving clozapine had schizophrenia compared to less than 50 % of those receiving other second-generation antipsychotics. Overall, 109 patients (52 %) received off-label prescriptions of second-generation antipsychotic drugs. This survey indicates that clozapine was mostly reserved for severe cases and poor responders; the high rate of off-label prescriptions highlights the gap existing between recommendations derived from randomised clinical trials and the current use of drugs.

Introduction

In recent years, the availability of second-generation antipsychotics (APs) has undoubtedly extended the treatment options available to patients suffering from schizophrenia [4]. Although the more treatment options there are, the greater choice one should have, the use of these new compounds in clinical practice has generated some concern.

The literature has consistently shown that clozapine is more effective than typical drugs in refractory schizophrenia [14] [2]. However, there is some concern about how subjects with refractory schizophrenia are identified [5]. Research criteria vary across studies, from stringent definitions of refractoriness (a minimum of three six-week trials in the preceding five years with typical APs from at least two chemical classes at doses equivalent to at least 1000 mg/day of chlorpromazine with no clinical response) to less stringent definitions of partial responsiveness, to typical APs [10]. These criteria have always been applied in well-defined, selected patient populations and treatment settings; in clinical practice, however, it might be harder to use them to establish refractoriness. In Italy, clozapine can only be prescribed to patients suffering from resistant schizophrenia, that is, patients who did not respond, or suffer intolerable side-effects, from at least two typical APs prescribed for at least six weeks and belonging to different pharmacological classes.

Besides clozapine, there is concern about the clinical use of risperidone, olanzapine and quetiapine. Whether or not these should be prescribed as first-line compounds in the pharmacological management of schizophrenia or whether they should be used only when typical APs have failed is subject to debate [7] [8]. Systematic reviews and clinical guidelines provide discordant recommendations. According to the Texas Medication Algorithm Project, for example, older APs are relegated to a fourth-line position after risperidone, olanzapine and quetiapine [11]; but in contrast, there is no clear evidence that second-generation APs are more effective or better tolerated than conventional compounds according to a more recent meta-analysis [6]. In Italy, the Italian drug reimbursement committee stated that olanzapine, risperidone and quetiapine can be prescribed to patients suffering from either schizophrenia or resistant schizophrenia. However, the National Health System provides full reimbursement only when prescribed in resistant schizophrenia [1].

In addition to patients suffering from schizophrenia, patients with schizophrenia-related disorders, bipolar affective disorders or schizoaffective disorders, or those with behavioural symptoms associated with personality disorders or dementia or mental retardation may receive antipsychotic drugs in clinical practice. These prescriptions are off-label in most European and non-European countries. Data collected in Europe suggested that old and new APs are very frequently prescribed for off-label indications [9] [15].

The present study was carried out to investigate the routine use of second-generation APs in the Italian psychiatric care system. We tested the hypothesis that patients receiving clozapine were more severely ill than those receiving any other second-generation AP, and we estimated the proportion of patients receiving off-label prescribing.

Methods

Study Area

Outpatient psychiatric services in the Milan area are public agencies which provide psychiatric care to 160 000 residents in an urban and suburban area. These community services are part of departments comprising psychiatric wards in the general hospital, psychiatric residential rehabilitative centres and staffed or unstaffed apartments. Outpatient psychiatric services represent the operational units in charge of managing all psychiatric services provided to patients belonging to their catchment areas. All but emergency cases are expected to keep their first contact with these facilities. They are also charged to act as the psychiatric interface of the network of general practitioners providing primary general care to all residents.

Study population

Over a six-month interval, in the second half of the year 2000, seven outpatient psychiatric services in the Milan area enrolled a consecutive case series of patients who were being treated, or had started treatment, with one of the second-generation APs - clozapine, olanzapine, risperidone or quetiapine. For each patient, the treating psychiatrist collected information on sociodemographic and clinical variables, diagnosis according to ICD-X criteria, current psychotropic drug use and dose regimens, side effects and past use of typical AP drugs. In addition, information on symptoms and behavioural problems, problems with relationships, activities of daily life, living conditions and at work were collected using the Health of the Nation Outcome Scale (HoNOS), a simple tool that has been found useful as a present state profile in the context of everyday clinical practice [3]. The HoNOS is a 12-item scale that measures the following four dimensions: behavioural problems (overactivity and aggressiveness, non-accidental self-injury, problem-drinking or drug taking); disability problems (cognitive problems, physical illness, other disability problems); psychiatric symptoms (hallucinations, delusions, depressed mood, other symptoms); social relationship and support (problems with relationships, with activities of daily life, with living conditions and problems with occupation). For each item, the score ranges from 0 (best health) to 4 (worst health) [13].

If olanzapine, risperidone and quetiapine were prescribed to patients not suffering from schizophrenia, their use was defined as off-label. According to the Italian National Formulary, if clozapine was prescribed to patients not suffering from resistant schizophrenia, that is, patients who did not respond, or suffered intolerable side-effects to at least two typical APs prescribed for at least six weeks and belonging to different pharmacological classes, its use was defined as off-label.

Daily AP doses in milligrams were converted into multiples of the Defined Daily Dose (DDD) for each drug by dividing the prescribed daily dose by the DDD (daily dose/DDD) [16]. This gave the number of DDDs of APs prescribed at study entry. A ratio of one indicates that the dose prescribed is equal to the DDD of that drug; a ratio greater than one indicates an excessive dosage, while a ratio lower than one means a low dose.

Data are expressed as number and percentage of the overall population. Categorical data were analysed by the chi-squared test; the Mann-Whitney two-sample analysis was used to analyse continuous non-normal data. Statistical analyses were done with STATA 4.0 [12].

Results

Differences between patients treated with clozapine and patients treated with other second-generation APs

A total of 209 patients were enrolled. More than a third of the patients (70 patients) were treated with risperidone, about a third with olanzapine (61 patients), a third with clozapine (62 patients) and less than 10 % with quetiapine (19 patients). Three patients received two second-generation compounds concurrently. No difference emerged between clozapine-treated patients and those receiving other second-generation APs in terms of sex and age distribution; however, more clozapine-treated patients were living in residential facilities, were unemployed and had a long psychiatric history (Table [1]).

Nearly 80 % of patients receiving clozapine were suffering from schizophrenia compared to less than 50 % of those receiving other second-generation APs. More than half the clozapine patients had already been treated with two or more typical APs compared to less than half of those receiving any other second-generation compound (Table [1]).

Table [2] presents the pharmacoepidemiology of psychotropic drugs. Length of current therapy with second-generation APs was similar for clozapine-treated patients and those receiving other second-generation drugs, while the association of a second-generation APs and benzodiazepines was less frequent for clozapine-treated patients than for patients receiving any other second-generation compound. Table [2] presents the median and mean maintenance doses for clozapine, olanzapine, risperidone and quetiapine. The mean DDD of APs received by clozapine-treated patients (clozapine + typical drugs) was 1.10 (SD 0.50), while the mean DDD of APs received by patients treated with any other second-generation compound (second-generation AP + typical drug) was 0.95 (0.71), indicating higher overall dose regimens in clozapine-treated patients (z-statistic 2.79, p = .005) (Fig. [1]).

The mean HoNOS score was similar in the two study populations; however, comparison of the HoNOS sub-scales showed that clozapine patients had more problems with activities of daily life and with living conditions (Table [3]).

Off-label prescribing of second-generation antipsychotics

Overall, 109 patients (52 %) received off-label prescriptions of second-generation APs. Around 30 % of clozapine-treated patients were off-label because they did not meet the refractoriness criteria; more than 10 % because they did not meet the refractoriness criteria and the diagnostic criteria, and less than 10 % because they did not meet the diagnostic criteria (Fig. [2]). Patients who were prescribed olanzapine, risperidone and quetiapine off-label ranged between 51 % and 65 % (Fig. [2]).

Discussion

The present survey indicates that patients using clozapine differ from those receiving other second-generation APs. Differences refer to diagnosis, living conditions, length of psychiatric history, number of typical drugs previously received, and overall doses currently received. From a social and functional perspective, subjects receiving clozapine had more problems with activities of daily living and with living conditions. Overall, it seems that clozapine was mostly reserved for severe cases and poor responders. Similar findings emerged from a survey conducted in a south London service on a sample of patients receiving second-generation APs [5]. Patients on clozapine were mostly male, had more than 15 years of contact with psychiatric services, and were poor responders. In contrast, patients prescribed other second-generation compounds had less than five years of contact with psychiatric services and had responded earlier to typical drugs, suggesting that second-generation APs are replacing older compounds in the first-line treatment of schizophrenia, at least in south London. In our sample, nearly 60 % of clozapine-treated patients had already been treated with at least two typical compounds, and clozapine was the first-line drug in only one case, given together with risperidone. Most patients receiving other second-generation APs had previously been treated with typical compounds, although 20 % received the new compounds as first-line therapy. This figure has to be interpreted carefully; in Italy, patients have to pay the full prize if olanzapine, risperidone and quetiapine are prescribed as first-line treatments. In many cases, this might have limited their use to patients who had already been treated with typical APs. However, this policy has recently been changed, and from 2001 onwards, olanzapine, risperidone and quetiapine are fully reimbursed to all patients suffering from schizophrenia. The expected result is that the evolving standard of care will lead to an increasing proportion of patients receiving second-generation APs as first-line therapy.

In more than fifty percent of patients receiving second-generation APs, prescriptions were for off-label indications. Lowe-Ponsford and Baldwing estimated in a sample of UK psychiatrists that 65 % had prescribed medication off-label within the past month [9], while Weiss et al. showed that in 66.5 % the drug was prescribed for off-label indications in a sample of Austrian patients treated with typical compounds [15]. In the Austrian survey, a medication was not considered off-label when it was prescribed for the treatment of schizophrenia, schizophreniform disorder, mania, or schizoaffective disorder, the so-called ”classical indications” for AP treatment. Therefore, Weiss’s criteria were less stringent than those adopted in the present survey, which considered off-label any prescriptions in patients without a diagnosis of schizophrenia. We adopted this criterion following the Italian National Formulary, although labels for olanzapine, risperidone and quetiapine are ambiguous and differ from each other. According to the Italian National Formulary, olanzapine is ”indicated in the treatment of schizophrenia”, risperidone is ”indicated in the treatment of acute and chronic schizophrenic psychoses”, and quetiapine ”is indicated in the treatment of acute and chronic psychosis, including schizophrenia”. It would probably be important to try and use similar labels for compounds indicated for the treatment of similar conditions, and to state clearly whether a compound is indicated for the treatment of the whole spectrum of psychoses, in bipolar affective illness, or in other disorders.

The situation is different for clozapine. Around 40 % of cases received a prescription off-label because patients did not fulfil the Italian National Formulary criteria for refractoriness. Frangou and Lewis showed that clinician’s definitions of resistance to treatment varied in routine clinical practice in London, lying below the threshold used in research [5]. We adopted a stringent definition of refractoriness in the sense that when the treating psychiatrists were not sure whether patients had been treated with at least two typical compounds - because the patient’s history was not entirely known - their prescriptions were classified as off-label.

Obviously, it cannot be assumed that all patients given off-label clozapine or that other second-generation APs were irrationally prescribed. However, this proportion of patients highlights the gap existing between recommendations derived from randomised clinical trials and the current use of drugs. From a clinician’s perspective, it would be of interest to recognise and follow these patients. Clinical databases could be developed to monitor the long-term outcome of all subjects receiving off-label prescriptions; these patients, at the moment, are an ”orphan” population since trials have rarely enrolled and followed patients not fulfilling precise diagnostic criteria, stringent selection criteria or stringent refractoriness criteria. Nevertheless, these subjects account for an epidemiologically important proportion of typical patients followed every day in typical settings.

References

• 1 Barbui C, Campomori A, Mezzalira L, Da Cas R, Garattini S. Psychotropic drug use in Italy 1984 - 1999: the impact of a change in reimbursement status.  Int Clin. Psychopharmacol 2001;  16 227-233
  PubMedSearch in Google Scholar
• 2 Chakos M, Lieberman J, Hoffman E, Bradford D, Sheitman B. Effectiveness of second-generation antipsychotics in patients with treatment-resistant schizophrenia: a review and meta-analysis of randomised trials.  Am J Psychiatry. 2001;  158 518-526
  Search in Google Scholar
• 3 Curtis R, Beevor A. Health of the Nation Outcome Scales. In Wing JK, editor Measurement for Mental Health. London; College Research Unit 1995
  Search in Google Scholar
• 4 Fleischhacker W W. Pharmacological treatment of schizophrenia: a review. In Maj M, Sartorius N, editors Schizophrenia. Vol. 2. WPA Series: Evidence and Experience in Psychiatry. John Wiley & Sons 1999: 75-107
  Search in Google Scholar
• 5 Frangou S, Lewis M. Atypical antipsychotics in ordinary clinical practice: a pharmaco-epidemiologic survey in a south London service.  Eur Psychiatry. 2000;  15 220-226
  CrossrefPubMedSearch in Google Scholar
• 6 Geddes J, Freemantle N, Harrison P, Bebbington P for the National Schizophrenia Guideline Development G roup. Atypical antipsychotics in the treatment of schizophrenia: systematic overview and meta-regression analysis.  BMJ. 2000;  321 1371-1376
  CrossrefPubMedSearch in Google Scholar
• 7 Kapur S, Remington G. Atypical antipsychotics. Patients value the lower incidence of extrapyramidal side effects.  BMJ. 2000;  321 1360-1361
  CrossrefPubMedSearch in Google Scholar
• 8 Leucht S, Pitschel-Walz G, Abraham D, Kissling W. Efficacy and extrapyramidal side-effects of the new antipsychotics olanzapine, quetiapine, risperidone, and sertindole compared to conventional antipsychotics and placebo. A meta-analysis of randomised controlled trials.  Schizophr Res. 1999;  35 51-68
  CrossrefPubMedSearch in Google Scholar
• 9 Lowe-Ponsford F, Baldwin D. Off-label prescribing by psychiatrists.  Psychiatric Bull. 2000;  24 415-417
  PubMedSearch in Google Scholar
• 10 Meltzer H Y. Treatment-resistant schizophrenia - the role of clozapine.  Curr Med Res Opinion. 1997;  14 1-20
  PubMedSearch in Google Scholar
• 11 Miller A L, Chiles J A, Chiles J K. et al . Texas Medication Algorithm Project (TMAP) schizophrenia algorithms.  J Clin Psychiatry. 1999;  60 649-657
  PubMedSearch in Google Scholar
• 12 StataCorp. Stata Statistical Software: Release 4. College Station, TX: Stata Corporation. 1995
  Search in Google Scholar
• 13 Stein G S. Usefulness of the Health of the Nation Outcome Scales.  Br J Psychiatry. 1999;  174 375-377
  PubMedSearch in Google Scholar
• 14 Wahlbeck K, Cheine M, Essali A, Adams C. Evidence of clozapine’s effectiveness in schizophrenia: a systematic review and meta-analysis of randomised trials.  Am J Psychiatry. 1999;  156 990-999
  PubMedSearch in Google Scholar
• 15 Weiss E, Hummer M, Koller D, Ulmer H, Fleischhacker W W. Off-label use of antipsychotic drugs.  J Clin Psychopharmacol. 2000;  20 695-698
  CrossrefPubMedSearch in Google Scholar
• 16 WHO Collaborating Centre for Drug Statistic Methodology. Guidelines for ATC Classification and DDD Assignment. WHO Oslo; 1996
  Search in Google Scholar

Dr. Corrado Barbui

Dept of Medicine and Public Health

Psychiatry Section

University of Verona

Ospedale Policlinico

37134 Verona

Italy

Phone: +39 (045) 807 44 41

Fax: +39 (045) 585 871

Email: corrado.barbui@univr.it

References

• 1 Barbui C, Campomori A, Mezzalira L, Da Cas R, Garattini S. Psychotropic drug use in Italy 1984 - 1999: the impact of a change in reimbursement status.  Int Clin. Psychopharmacol 2001;  16 227-233
  PubMedSearch in Google Scholar
• 2 Chakos M, Lieberman J, Hoffman E, Bradford D, Sheitman B. Effectiveness of second-generation antipsychotics in patients with treatment-resistant schizophrenia: a review and meta-analysis of randomised trials.  Am J Psychiatry. 2001;  158 518-526
  Search in Google Scholar
• 3 Curtis R, Beevor A. Health of the Nation Outcome Scales. In Wing JK, editor Measurement for Mental Health. London; College Research Unit 1995
  Search in Google Scholar
• 4 Fleischhacker W W. Pharmacological treatment of schizophrenia: a review. In Maj M, Sartorius N, editors Schizophrenia. Vol. 2. WPA Series: Evidence and Experience in Psychiatry. John Wiley & Sons 1999: 75-107
  Search in Google Scholar
• 5 Frangou S, Lewis M. Atypical antipsychotics in ordinary clinical practice: a pharmaco-epidemiologic survey in a south London service.  Eur Psychiatry. 2000;  15 220-226
  CrossrefPubMedSearch in Google Scholar
• 6 Geddes J, Freemantle N, Harrison P, Bebbington P for the National Schizophrenia Guideline Development G roup. Atypical antipsychotics in the treatment of schizophrenia: systematic overview and meta-regression analysis.  BMJ. 2000;  321 1371-1376
  CrossrefPubMedSearch in Google Scholar
• 7 Kapur S, Remington G. Atypical antipsychotics. Patients value the lower incidence of extrapyramidal side effects.  BMJ. 2000;  321 1360-1361
  CrossrefPubMedSearch in Google Scholar
• 8 Leucht S, Pitschel-Walz G, Abraham D, Kissling W. Efficacy and extrapyramidal side-effects of the new antipsychotics olanzapine, quetiapine, risperidone, and sertindole compared to conventional antipsychotics and placebo. A meta-analysis of randomised controlled trials.  Schizophr Res. 1999;  35 51-68
  CrossrefPubMedSearch in Google Scholar
• 9 Lowe-Ponsford F, Baldwin D. Off-label prescribing by psychiatrists.  Psychiatric Bull. 2000;  24 415-417
  PubMedSearch in Google Scholar
• 10 Meltzer H Y. Treatment-resistant schizophrenia - the role of clozapine.  Curr Med Res Opinion. 1997;  14 1-20
  PubMedSearch in Google Scholar
• 11 Miller A L, Chiles J A, Chiles J K. et al . Texas Medication Algorithm Project (TMAP) schizophrenia algorithms.  J Clin Psychiatry. 1999;  60 649-657
  PubMedSearch in Google Scholar
• 12 StataCorp. Stata Statistical Software: Release 4. College Station, TX: Stata Corporation. 1995
  Search in Google Scholar
• 13 Stein G S. Usefulness of the Health of the Nation Outcome Scales.  Br J Psychiatry. 1999;  174 375-377
  PubMedSearch in Google Scholar
• 14 Wahlbeck K, Cheine M, Essali A, Adams C. Evidence of clozapine’s effectiveness in schizophrenia: a systematic review and meta-analysis of randomised trials.  Am J Psychiatry. 1999;  156 990-999
  PubMedSearch in Google Scholar
• 15 Weiss E, Hummer M, Koller D, Ulmer H, Fleischhacker W W. Off-label use of antipsychotic drugs.  J Clin Psychopharmacol. 2000;  20 695-698
  CrossrefPubMedSearch in Google Scholar
• 16 WHO Collaborating Centre for Drug Statistic Methodology. Guidelines for ATC Classification and DDD Assignment. WHO Oslo; 1996
  Search in Google Scholar

Dr. Corrado Barbui

Dept of Medicine and Public Health

Psychiatry Section

University of Verona

Ospedale Policlinico

37134 Verona

Italy

Phone: +39 (045) 807 44 41

Fax: +39 (045) 585 871

Email: corrado.barbui@univr.it