|Home | About | Journals | Submit | Contact Us | Français|
To describe the pattern of antipsychotic drug prescribing in patients with first episode psychosis, with more emphasis in the use of clozapine in this group of patients.
A cross-sectional survey involving six early intervention service (EIS) teams in the West Midlands was conducted. Data was extracted from case notes and electronic records by clinicians working in each participating team. The pattern of antipsychotic prescribing and the changes that took place after being accepted in EIS, including the use of clozapine, was established. Clinicians involved in the treatment of patients in each team rated the overall clinical response to treatment based on the presence or absence of positive psychotic symptoms.
431 patients with FEP were included in the final analysis. Low antipsychotic discontinuation rate was observed, with the majority (88.2%) still being prescribed antipsychotics. Most (77.3%) were prescribed second-generation antipsychotic drugs, with olanzapine (21.8%) and aripiprazole (19.7%) being the most frequently prescribed antipsychotics. There was low rate use of antipsychotic combinations (7.4%), high dose antipsychotic regime (3.9%), low depot antipsychotic prescribing (9.3%), and clozapine use was low (9.7%). On average, three antipsychotics were tried before clozapine was initiated and it took on average 19.5 months from being accepted into EIS to clozapine being initiated.
The majority of patients were prescribed antipsychotics within the guidelines. EIS was associated with an overall low antipsychotic discontinuation. There was also a short waiting time before clozapine was initiated following patients being accepted into EIS.
Early intervention services (EISs) aim to treat those with first-episode psychosis (FEP) at the earliest possible opportunity. Delayed treatment of FEP is associated with poor outcome [Johnstone et al. 1986; Wyatt, 1991]. Early treatment of patients with FEP is important given the observation that patients with FEP have an increased risk of suicide, suicide attempts [Harvey et al. 2008] and increased substance use [Hides et al. 2006; Koskinen et al. 2010].
Antipsychotic drugs remain the cornerstone in the treatment of FEP. As to the choice of antipsychotics, many guidelines do not provide an algorithm of what antipsychotics or group of antipsychotics are to be used first. Some advise that first-generation antipsychotics (FGAs) and second-generation antipsychotics (SGAs) can be equally utilized as long as they are tolerated [National Institute for Health and Care Excellence (NICE), 2014; Scottish Intercollegiate Guidelines Network, 2013; Barnes and Schizophrenia Consensus Group of British Association for Psychopharmacology, 2011]. Other guidelines, recommend SGAs over FGAs [Lehman et al. 2004; Canadian Psychiatric Association, 2005]; while others specifically mention drugs such as olanzapine and ziprasidone be avoided due to metabolic syndromes and cardiac concerns respectively [Buchanan et al. 2010; Marder et al. 2002].
It is not clear how these guidelines are used in clinical practice. The pattern of antipsychotic use in patients with multiple episodes of psychosis has been extensively studied. However, this has not been the case for patients with FEP in EISs and is more apparent when it comes to the use of clozapine within this group of patients [Vera et al. 2012; Zhang et al. 2013]. The two common findings about the use of antipsychotics in clinical practice are a high prevalence of polypharmacy and a significant delay in starting clozapine [Taylor et al. 2003; Tungaraza et al. 2010].
It is well known that patients with FEP have a high response rate to antipsychotic medication. It is estimated that between 40–90% may respond to antipsychotic medication [Robinson et al. 2005]. It has been shown however, that failure to respond to the first antipsychotic prescribed in FEP is associated with a poor response to subsequent nonclozapine antipsychotics. A study by Agid and colleagues noted that while 76% of their study group responded on the first antipsychotic, only 23% responded on a trial of a second antipsychotic medication [Agid et al. 2007]. Clozapine is reserved as a third line treatment option for those who have not responded to two antipsychotics where at least one is from a different group. NICE guidance advises to ‘offer clozapine to people with schizophrenia whose illness has not responded adequately to treatment despite the sequential use of adequate doses of at least two different antipsychotic drugs. At least one of the drugs should be a nonclozapine second-generation antipsychotic’ [NICE, 2014]. We therefore aim to describe the pattern of use of antipsychotics in EISs. We specifically aim to describe the use of clozapine relative to other antipsychotic prescribing in this population.
In 2011, 10 EIS teams in the West Midlands, UK and surrounding areas were invited to participate in a survey; 6 teams agreed to take part. Approval was sought from individual trusts that took part.
One proforma to be used to extract information from case notes and electronic records across all teams was developed. Treating clinicians working in each EIS extracted the information on the following variables: diagnosis, age, sex, ethnicity, past history of hospital admission, antipsychotic use including clozapine in chronological order since patients were accepted in EIS. Clinicians in each team prescribed and changed antipsychotics depending on the clinical need of each patient.
For the purpose of this survey, each time an antipsychotic was changed, that was considered as a trial of an antipsychotic drug. Depending on the number of changes that took place, one could have had several trials of antipsychotics (both orally given and as depot medications). Clinicians in each team rated the clinical improvement based on the presence or absence of positive psychotic symptoms. All patients with a diagnosis of FEP, as long as they were or have been on antipsychotics, were included in the study. Those excluded from the study included those with a diagnosis of FEP but had not yet been started on antipsychotics and those with diagnoses other than FEP.
Data were analysed using SPSS version 21.0 (IBM Corp 2012). Descriptive statistics were used to describe and Chi-square tests were used to investigate the association between variables of interest.
A total of 470 individuals were being treated within the six EIS teams at the time of the survey. A total of 39 individuals were excluded for not meeting the inclusion criteria or because a significant amount of information was missing. The final analysis comprised of 431 patients (see Table 1 for demographic characteristics of the group). A total of 302 patients (70.1%) were male; the group mean age was 24.8 (range 16–37). Overall, 266 patients (61.7%), were White, and 38.3% were from black and minority ethnic groups of which 48 (11.1%) were from a black ethnic minority. A total of 205 patients (47.6%) had a history of hospital admission in the past. No association was observed between hospitalization and ethnicity [X2 = 7.79, degrees of freedom (df) = 6, p = 0.254]. The duration of being in EISs varied widely from 1–82 months (mean 22.27, median 19).
A total of 380 patients (88.2%) were still being prescribed antipsychotics and only 51 (11.8%) had discontinued their medication (see Figure 1). Overall, 333 (77.3%) were prescribed SGA monotherapy. A total of 32 patients (7.4%) were receiving a combination of antipsychotic drugs either as atypical/atypical or typical/atypical, the former being the most common one. For those on combinations, no-one was prescribed more than two antipsychotics. A total of 16 different types of combinations were prescribed; the common ones being quetiapine and aripiprazole (7); risperdal consta and oral risperidone (6); clozapine and aripiprazole (3); clozapine and amisulpiride (2); and olanzapine and aripiprazole (2). Overall, 317 patients (about 74%) had tried no more than two antipsychotics, while 114 (26%) had a trial of more than two. The majority of patients however, (363, 84.2%) were prescribed antipsychotics within the British National Formulary (BNF) limit and 17 patients (3.9) had received antipsychotic doses above BNF recommendations.
Symptom control was achieved in 212 patients (49.2%) who were classified as having no positive symptoms. A total of 194 patients (45%) were still experiencing positive symptoms, whereas 25 (5.8%) could not be classified. Having ongoing symptoms was associated with being prescribed a high dose (Yates X2 = 8.34, df = 1, p = 0.0039), and having tried more than two antipsychotics (Pearson X2 = 28.76, df = 2, p < 0.001) (see Figure 2). Symptom resolution was associated with the female sex (Yates X2 = 10.89, df = 1, p = 0.001). There was no association of having ongoing symptoms and a history of hospital admission (Yates X2 = 0.02, df = 1, p = 0.88).
At the time of conducting the survey the most prescribed antipsychotic drugs (as monotherapy) were olanzapine (94, 21.8%), aripiprazole (85, 19.7%), quetiapine (65, 15.1%), risperidone (40, 9.3%), clozapine (23, 5.3%), risperdal consta (15, 3.5%), and flupenthixol decanoate (8, 1.9%). However, the most frequently prescribed antipsychotics varied from time to time as different antipsychotics were tried when changes to medication were undertaken (see Table 2). Olanzapine (152, 35%) was the most prescribed drug at the beginning of treatment (first antipsychotic) followed by risperidone (101, 23.4%) and then quetiapine (88, 20.4%). On the trial of a second antipsychotic drug, aripiprazole (80, 34%) was the most preferred followed by olanzapine (47, 20%). On considering the third antipsychotic option, quetiapine (17, 14.9%) was number one, followed closely by olanzapine and aripiprazole, each with 16 (14.0%). Clozapine came out on top when changing antipsychotic for the fourth time, with 9 individuals being tried on it.
Between 1–7 different antipsychotics (median, 2) were tried. A total of 196 patients (45.5%) had tried one antipsychotic drug (27 stopped medication at that time), 121 (28.1%) had tried two antipsychotics (14 stopped their medication), 59 (13.7%) had tried 3 antipsychotic drugs and the remaining 55 (12.8%) had tried between 4–7 antipsychotics. Overall, 169 (39.2%) were still continuing on the first antipsychotic drug that was prescribed. Among the 51 who had stopped taking antipsychotics, 41 (80.4%) did so at the time when a second antipsychotic was being tried.
A total of 61 (14.2%) had tried depot medication during their period in EISs and 40 (9.3%) were still being prescribed depot formulations. The most commonly prescribed depot medications were risperdal consta (25), flupenthixol decanoate (8), zuclopenthixol decanoate (3), and piportil palmitate (3). Of the 61 that had tried depot medication, 16 (26.2%) did so on the trial of a second antipsychotic and 23 (38%) on the trial of a third antipsychotic drug. Depot medications were prescribed as a first option to 7 (11.5%) patients but were given as the fourth option to 8 (13.1%) patients.
A total of 33 individuals (9.7%) out of 339 with nonaffective psychosis had a trial of clozapine. Overall, 28 (8.3%) were still taking clozapine at the time of the study. A total of 27 patients (81.8%) were male, 22 (66.7%) were White, 3 (9.1%) were black, and 8 (24.2%) were Asian. Overall, 24 patients (72.7%) had a period of hospital admission. Their length of stay in EISs varied from 5–49 months (mean 28.91, median 29). Mean age was 24.91 (range 18–34), and they had tried between 1–5 antipsychotics (median, 3) before clozapine was started.
Among the 33 who had tried clozapine, 26 (78.8%) were started in EISs, and the remaining 7 (21.2%) were started elsewhere. The 26, for whom clozapine was started within EISs, had been in the service for a period ranging from 5–49 months (mean 30.62, median 30). The time it took to prescribe clozapine after being in contact with EISs varied between 2–37 months (mean 19.5).
Of those who were still prescribed clozapine (n = 28), 23 (82.1%) were on clozapine monotherapy while 5 were on combinations. A total of 3 individuals had clozapine combined with aripiprazole and 2 with amisulpiride. All were prescribed antipsychotics within the BNF limit. Interest-ingly, all except 3 (10.7%) patients were classified as still having positive psychotic symptoms. There was association of having positive symptoms and taking clozapine (Yates X2 = 18.24, df = 1, p < 0.001).
This study explored the prescribing pattern of antipsychotics including clozapine in EISs. Our main findings can be summarized as follows. The majority of patients (70.1%) were male; nearly half (47.6%) had a period of hospital admission in the past. Interestingly there was low discontinuation rate (11.8%) and 380 (88.2%) were still being prescribed antipsychotics. The majority (77.3%) of patients were prescribed SGA drugs and most had antipsychotic doses within the BNF limit. A total of 50% were classified as having no positive symptoms. However, 45% were still experiencing minor or significant positive symptoms. Nearly 40% were still taking the first prescribed antipsychotic drug. Overall, 80% of those who stopped their antipsychotics did so while they were on a second antipsychotic drug trial. About 8.3% of those suffering from nonaffective psychosis were still taking clozapine. Clozapine was initiated on average within 19.5 months of being accepted in EISs.
A number of factors contribute to the strength of our study. We observed what actually took place on a day-to-day clinical setting, providing a naturalistic observation. A large sample size was gathered, few patients were excluded, and the involvement of a number of teams made it possible to generalize to other EIS teams in the UK. Data were collected by clinicians working in these teams and were familiar with the patients, their symptoms, and their progress and were thus able to provide a more meaningful and realistic clinical picture.
However, the survey looked at what was prescribed from the time patients joined the EIS and may have not captured what took place prior to their being accepted in the EIS. As an example, seven patients had clozapine started elsewhere before joining the EIS. This may also apply to other antipsychotics. However, for the majority of the patients the pattern of prescribing was available. The outcome measure of symptom improvement was based on clinical judgement.
It is of interest to note that 70% of our sample was male. This appears to be typical of EIS samples. The reasons behind the sex difference in the presentation in FEP and schizophrenia continue to be debated. It is well known that in FEP, males present early with psychosis between 18–25 years versus female 25–35 and this may explain the observed difference [Ochoa et al. 2012]. A number of reviews [Ochoa et al. 2012; Abel et al. 2010] have noted that even though epidemiological studies appear to show that there is increased incidence of schizophrenia in men, population-based studies indicate that the prevalence is the same between men and women. A review of sex difference in non-epidemiological studies [Longenecker et al. 2010] noted that men are over represented in study samples and they can make up to 66% of the sample population, as was the case in our study. The underlying reasons for men being overrepresented in non-epidemiological samples remains speculative but has included the early age of onset in men, women presenting with depressive symptoms, differing diagnostic criteria, and availability of and access to treatment [Abel et al. 2010; Longenecker et al. 2010; Ochoa et al. 2012].
Most of our patients were prescribed one antipsychotic drug within the BNF limit with only 17 (3.9%) being prescribed antipsychotic doses above the BNF limit and 32, (7.5%) being prescribed combinations of antipsychotic drugs. A recent FEP study in the United States (US) identified a group of patients, among whom 8.8% were prescribed higher than recommended doses of antipsychotics [Robinson et al. 2015]. The same study observed that 23.3% were prescribed combinations of antipsychotic drugs. We noted that those receiving combinations and those who have tried more than two antipsychotics were more likely to be classified as having positive symptoms. This may indicate that they were either experiencing severe illness or having a poor response to antipsychotics.
Overall, the continuation with antipsychotic treatment as a whole was impressive with only 11.8% discontinuing their antipsychotics. A total of 80% of those who discontinued their medication did so by the time a second antipsychotic drug was tried. A European open, randomized, controlled trial involving patients with FEP in 14 countries observed that 42% discontinued antipsychotics within 12 months [Kahn et al. 2008]. A large Swedish study of first time hospitalized patients with schizophrenia observed that 54.3% did not collect their prescription within 30 days of discharge and on a 2-year follow-up period, 34.3% discontinued the use of antipsychotics [Tiihonen et al. 2011]. Some of the key features of good EISs include intensive follow up, a small case load, family involvement, the use of SGAs that provide simplified medication regime, and shared decision-making. These factors have been observed to enhance adherence and could have contributed to the observed high adherence in our study [Haddad et al. 2014]. An alternative explanation is that this study reflects the outcome of patients who were willing to engage with the services, whereas those with poor compliance and poor engagement may have lost contact with the EISs and would therefore not be represented.
Our study also highlights the need for clinicians to work closely with patients and carers when the choice of the first two antipsychotics are being considered given the high discontinuation rate we have observed around that time. It is therefore very important that the choice of the first two antipsychotics be carefully and seriously considered in terms of side effect profile (including metabolic effects) and its known effectiveness, in collaboration with patients and carers, to reduce the high discontinuation rate around that period.
Despite the low discontinuation rate, 50% were rated as having no positive symptoms. These rates are in broad agreement with other studies. A 7-year follow-up study in Australia for patients in early intervention, Early Psychosis Prevention and Intervention Centre, noted that symptomatic remission at follow up was observed in 37–59% of their various subgroups [Henry et al. 2010]. Another study noted that after 5 years of treatment, 47.2% of FEP patients were in remission after 2 years of treatment [Robinson et al. 2005]. This highlights the need for implementing early use of clozapine and also the combination of pharmacological and psychosocial interventions to achieve remission.
Just over a quarter of our group had tried 3–7 different antipsychotics. Existing guidelines recommend clozapine be considered when two antipsychotics have failed. However, we noted that there was low clozapine utilization. Clozapine underutilization is a worldwide problem and the reasons for its low use vary widely and include patients and clinicians concerns about clozapine side effects, and clinicians’ lack of knowledge and experience in its use [Tungaraza and Farooq, 2015]. The rate of 8.3% in our sample is higher than that was observed in a US study (0.7%) among 300 patients with FEP who were prescribed antipsychotic monotherapy [Robinson et al. 2015]. This is, however, still lower than the observed prevalence of treatment-resistant schizophrenia as 26% of our group had tried more than two antipsychotics and so were eligible for clozapine trial. Clearly, more work needs to be done to improve clozapine uptake in FEP. Clozapine utilization in our study was much higher than that observed in a 16 multicentre FEP study in Spain, where 3.3% were prescribed the drug [Bioque et al. 2015]. A higher rate of clozapine use (12.5%) was noted in a sample of a Finnish study, among patients admitted first time with a diagnosis of schizophrenia [Tiihonen et al. 2011]. Despite this, our findings are encouraging considering that this was a first-episode population, an area where there is some reluctance among clinicians to use antipsychotics, let alone clozapine [Thompson et al. 2015].
Overall, there is lack of systematic studies in the use of clozapine in FEP. Studies that have tried to evaluate the effectiveness of clozapine as first line treatment in FEP indicate that though patients achieved remission of symptoms early on clozapine, the overall end result did not differ significantly from chlorpromazine or other FGAs [Girgis et al. 2011; Woerner et al. 2003] or SGAs [Vera et al. 2012; Robinson et al. 2005]. Very few randomized trials have been conducted to evaluate the use of clozapine in treatment-resistant FEP. In a naturalistic study involving 244 patients with FEP, a high response rate on clozapine (75%) was observed for those on whom clozapine was used after they had failed on two antipsychotics (i.e. olanzapine followed by risperidone or vice versa) [Agid et al. 2011].
Most of the current evidence of clozapine in patients with FEP is extrapolated from patients with multiple episodes. Since the seminal study by Kane and colleagues [Kane et al. 1988], it has been repeatedly shown in clinical trials [Lewis et al. 2006; Stroup et al. 2009] and in meta-analyses [Wahlbeck et al. 1999; Lieberman et al. 2005] that clozapine is effective in treating patients with treatment-resistant schizophrenia. Clozapine also confers a number of benefits when compared with other antipsychotics. It has been shown to be associated with low hospitalization rates [Duggan et al. 2003; Wheeler et al. 2009], reduced suicide and self-harm [Meltzer et al. 2003; Duggan et al. 2003], and reduced substance use [Drake et al. 2000; Brunette et al. 2006]; all these features being common among patients with FEP.
EISs appear to have reduced the duration it took to prescribe clozapine once patients were accepted into the service, as well as the number of antipsychotics used before clozapine was initiated. The duration of 19.5 months or 1.5 years after entry in EISs was noted before clozapine was prescribed compared with 4–9 years observed in those with multiple episodes [Howes et al. 2012; Taylor et al. 2003; Wheeler, 2008]. On average FEP patients had tried three antipsychotics before being prescribed clozapine; this was lower in comparison with some studies.
In conclusion, we observed that the overall antipsychotic prescribing pattern in patients with FEP was consistent with the guidelines for prescribing antipsychotics. The majority of patients were prescribed antipsychotic doses within the BNF limit. A low rate of antipsychotic discontinuation was noted too. However, there are still areas needing improvement considering that a significant number of patients had not achieved symptom remission, and clinicians were still trying other antipsychotics before considering clozapine, contrary to guidelines. The use of depot injections was also quite low. It is time that an algorithm-based approach be adopted in EISs for selecting and switching antipsychotics [Moore et al. 2007]; the aim at all stages being to achieve remission as defined by the standardized criteria [Andreasen et al. 2005].
Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Conflict of interest statement: The authors declare that there is no conflict of interest.
Tongeji E. Tungaraza, Consultant General Adult Psychiatrist (Rehabilitation), Partnerships in Care Ltd., Beverley House, 527–529 City Road, Birmingham, B617 8LL (Formerly Consultant EIS, Wolverhampton), UK.
Wakil Ahmed, Consultant General Adult Psychiatrist, MHHTT, Park House, North Manchester, Manchester Mental Health & Social Care Trust, UK.
Chinonyelum Chira, Consultant in Child and Adolescent Psychiatry, Birmingham Children’s Hospital, UK.
Erin Turner, Consultant Solihull Early Intervention Service, Birmingham and Solihull Mental Health Foundation Trust, UK.
Susan Mayaki, Consultant General Adult Psychiatrist, Assertive Outreach Team, and Rehabilitation, Hereford, UK.
Harpal Singh Nandhra, Consultant Psychiatrist, C&W Partnership Trust. Ashton house, Leamington Spa, UK.
Tom Edwards, Consultant Psychiatrist, Dudley and Walsall Mental Health Partnership NHS Trust, Walsall Assertive Outreach Team and Walsall North Community Recovery Service, Dorothy Pattison Hospital, Alumwell Close, Walsall, West Midlands, UK.
Saeed Farooq, Clinical Senior Lecturer, Research Institute for Primary Care & Health Sciences, Keele University, UK Honorary Consultant Psychiatrist, South Staffordshire and Shropshire NHS Foundation Trust, UK.