POMH-UK Topic 1 report 1b Prescribing of high-dose and combination antipsychotics on adult : Clinical introduction, methodology and glossary. March 2007 Prepared by the Prescribing Observatory for Mental Health. This report will be of interest to Executive Teams, Drugs & Therapeutics (D&T) Committees, Clinical Governance Committees, POMH Local Project Teams, clinicians, pharmacists and service user groups. Please use the following to cite this report: Prescribing Observatory for Mental Health (2007). Topic 1 report 1b. Prescribing of high-dose and combination antipsychotics on adult : 12-month re-audit
Introduction POMH-UK The Prescribing Observatory for Mental Health (POMH-UK) is a national quality improvement programme open to all specialist mental health services in the UK. POMH-UK works with mental health services to help improve prescribing practice in discrete areas ( Topics ) of prescribing practice. Topics involve audit cycles with supported change interventions. For this Topic, participating teams conducted a baseline audit of their practice in January 2006 and were offered a number of change interventions to support quality improvement. In January 2007 the same teams repeated the audit to measure changes in practice. Further analysis of your Trust data Ownership of all data submitted to POMH-UK is retained by the Trust that provided it. An Excel file containing all the data submitted by your Trust has been made available to your Trust s Local Project Team lead. Please contact this person if you wish to conduct further analyses of your Trust s data.
Clinical background The British National Formulary (BNF) provides recommended dose ranges for the licensed antipsychotics to treat schizophrenia; these can also be found in the Summary of Product Characteristics (SPC) for each drug at www.medicines.org.uk. These dose ranges are determined in relation to the efficacy and toxicity data for each drug. Clinical guidelines, such as those produced by NICE, recommend that, with a few exceptions, patients should be prescribed only one antipsychotic at a time. The evidence base for high-dose antipsychotics While there is considerable evidence for the effectiveness of antipsychotic drugs in the treatment of psychosis (Lehmann and Ban, 1997), there is no evidence to suggest that doses of antipsychotics higher than the recommended dosages are more effective than standard doses (Lehman et al, 1998). This also appears to hold true for cases where standard doses have failed to produce any benefit. The controlled studies comparing very high-doses of first-generation antipsychotics with standard dosage regimens for treatment resistant schizophrenia all failed to show a significant advantage for the high dosage (Thompson, 1994; Royal College of Psychiatrists, 2006). Further, higher doses have a greater risk of dose-related side effects. The evidence base for combined antipsychotics Sometimes people are prescribed more than one antipsychotic because they are switching from one antipsychotic to another, and there is cross-tapering of the doses of both drugs in the transition phase. Some people are prescribed a second antipsychotic drug PRN for the management of disturbed behaviour. For others, combined antipsychotics are prescribed because a patient s response to a single antipsychotic has proved less than satisfactory, and it is hoped that the addition of another will lead to an enhanced therapeutic effect. The effectiveness and side-effect burden associated with this approach have not been studied systematically in clinical trials (Freudenreich & Goff, 2002). The evidence that does exist suggests that the potential for harm may outweigh the potential for benefit; there is no convincing evidence that symptoms improve (Taylor et al, 2002; Centorrino et al, 2004) and patients who are prescribed combinations are more likely to receive a high total antipsychotic dose (Harrington et al, 2002a), experience side effects and spend longer in hospital (Centorrino et al, 2004). In the longer term, there is tentative evidence that mortality may be increased (Waddington et al, 1998). In 2006 the Medicines and Healthcare Regulatory Authority conducted a review of the cardiac safety of all antipsychotics available in the UK (MHRA, 2006). This led to the recommendation that the wording avoid concomitant neuroleptics should be added to the special warnings and precautions for use section of the summary of product characteristics (SPC; product licence) of every antipsychotic. For haloperidol, the most widely used as required antipsychotic
in the UK, the SPC also recommends that an ECG should be performed before treatment is started. Recent studies have found that approximately 10% of patients prescribed maintenance antipsychotic medication receive combined antipsychotics in the medium (at 3 months; Kreyenbuhl et al, 2006) and long term (1 year; Barbui et al, 2006). In around three- quarters of cases, these are combinations of FGAs and SGAs (Kreyenbuhl et al, 2006), which is likely to negate the main advantage of SGAs, i.e. their lower liability for extrapyramidal side effects (EPS). This notion is supported by the finding that patients prescribed combined SGAs (Carnahan et al, 2006) or combined SGAs and FGAs (Paton et al, 2003) are more likely to be prescribed medication to treat EPS than patients treated with a single SGA. Combined antipsychotics in patients with treatment-resistant schizophrenia For people with treatment-resistant schizophrenia who have shown a poor or only partial response to clozapine, the addition of either a first-generation or second-generation antipsychotic is a common clinical strategy. However, the research evidence to justify this practice is limited (Chong & Remington 2000), and the results from recent clinical trials are equivocal (Paton et al 2006). Summary It should be routine clinical practice to use one antipsychotic at a time, at a dose within the recommended range. Given the nature of the clinical evidence however, it is not possible to say that the risk-benefit balance would never favour prescription of an antipsychotic above the BNF recommended maximum dosage, or in combination with another, but prescribing in this way should always be part of an individual clinical trial as suggested by Stahl (2002), who stated: combined antipsychotics should only be considered following lack of response to multiple adequate trials of antipsychotic monotherapy, and then it should be administered as a time-limited trial that is closely monitored, and the combination should only be continued if there is evident therapeutic benefit. Findings from the POMH baseline audit 2006 In January 2006, 32 Trusts/healthcare organisations participated in the POMH baseline audit, submitting data for 3,492 patients from 218 wards. 36% of people in the total national sample were prescribed a total antipsychotic dose greater than 100% of the recommended maximum (range across participating Trusts 17-71%). 43% of people in the total national sample were prescribed more than one antipsychotic drug (range across participating Trusts 0-70%), the majority through the use of as required medication.
31% of people in the total national sample were prescribed a first and second generation antipsychotic in combination (range across participating Trusts 0-56%). The data were analysed to explore which factors influenced the prescription of combined and high-dose antipsychotics in people with schizophrenia or related disorders (ICD10 categories F20-29), the largest diagnostic group in the audit population. Of the 2,032 patients in this group, 1025 (50%) were prescribed combined antipsychotics and 863 (43%) were prescribed a high-dose. A binary logistic regression analysis was conducted with high-dose as the dependent variable, and combined antipsychotics and those variables that have previously been shown to be associated with high-dose (age band, ward type, MHA status and gender Lelliott et al, 2002) as the independent variables. Together, these variables correctly predicted the prescription of high-dose or not high-dose in 82% of cases. The only variable in this model that was significant at the level of P<0.001 was combined antipsychotics. A patient who was prescribed combined antipsychotics was more than 20 times more likely to be prescribed a high-dose as defined in our audit than a patient who was prescribed a single antipsychotic (Odds Ratio 23; 95% confidence intervals 18 to 29). The regression analysis was repeated without combined antipsychotics as an independent variable, in order to examine further the possible influence of the other variables. This model was relatively poor at predicting high-dose, yielding only 60% correct classifications. Adding ethnicity to the model did not improve its predictive power. The results of these analyses suggest that if age, gender, MHA status and ward type have any influence on prescribing practice with respect to high-dose and combined antipsychotics, it is modest. Therefore, the drivers for such prescribing must lie elsewhere, and the findings of the audit suggest that these are aspects of the clinical picture. The clinical teams participating in the audit reported that the main reasons for prescribing combined antipsychotics were disturbed behaviour and/or failure to respond to standard doses of antipsychotic monotherapy. This analysis was repeated on the re-audit data (see Section 1.3).
Audit standards Prescribing high-dose antipsychotics The maximum licensed dose for each antipsychotic drug is clearly outlined in its SPC, and in the BNF. The Royal College of Psychiatrists (2006) Council Report 138, a revised consensus statement on high-dose antipsychotic medication, concluded that current evidence did not justify the routine use of high-dose antipsychotic medication in general adult mental health services. Audit standard 1: The dose of an individual antipsychotic should be within its SPC/BNF limits. A high-dose of antipsychotic is defined here as a total daily dose (whether of a single antipsychotic or more than one prescribed in combination) greater than 100% of the maximum recommended daily dose. (Royal College of Psychiatrists, 2006). Prescribing combined antipsychotics NICE has published a Health Technology Appraisal (HTA) on the use of atypical antipsychotics in schizophrenia (NICE, 2002), and a treatment guideline for schizophrenia (NICE, 2003). Both the HTA guidance and the NICE treatment guideline contain audit standards that relate to the prescription of more than one antipsychotic drug simultaneously: Audit standard 2: Individuals receive only one antipsychotic at a time. This standard applies to 100% of individuals with schizophrenia. Exceptions: Individuals with schizophrenia who are receiving clozapine but who have not responded sufficiently; and individuals who are changing from one antipsychotic to another (NICE schizophrenia treatment guideline audit standard 6). Audit standard 3: First (typical) and second generation (atypical) antipsychotic drugs are not prescribed concurrently. This standard applies to 100% of individuals with schizophrenia. Exceptions: Any concurrent prescriptions are for a short period to cover changeover of medication. Local teams should agree on what constitutes a changeover period for audit purposes (HTA Audit standard 5).
Method The Prescribing Observatory for Mental Health (POMH-UK) invited all National Health Service (NHS) Trusts in the United Kingdom (UK) providing specialist mental health services to participate in a national audit of high-dose and combination antipsychotic prescribing. Each Trust that formally agreed to take part was asked to form a Local Project Team (LPT), with suggested membership of a psychiatrist, pharmacist, nurse, clinical governance staff member and at least two service users. Local Project Teams were invited to attend one of five regional introductory workshops to discuss and review the aims, objectives and methodology of the proposed audit. Comment and discussion at the workshops led to refinements of the audit methodology and data collection tool. The participating Trusts were self selected in that they chose to participate in the audit. All participating Trusts/organisations are listed in alphabetical order in Appendix B. Subjects and settings Each LPT was invited to include as many acute adult admission and psychiatric intensive care (PICU) wards as they wished. Teams were asked to submit data for all patients who, on a census day, occupied a bed on their selected wards and were being prescribed one or more antipsychotic drugs. To ease the burden of data collection, a different census day (within the period 9-22 January 2006 at baseline and 8-21 January 2007 at re-audit) could be chosen for each ward within the same Trust, but all data for any single ward had to be collected on the same census day. Data collected The following data were collected for all eligible patients on baseline and re-audit census days: Demographic variables (age, gender, ethnicity); Clinical variables (diagnostic grouping, Mental Health Act status); Names and dosage of all regular and PRN antipsychotic drugs prescribed on the census day. Note: For both regular and PRN drugs, the maximum dose that could be administered to a patient over a 24-hour period according to their prescription sheet was recorded, irrespective of whether it was administered or not. In addition, for patients who were prescribed more than one antipsychotic, the following was recorded: Primary reason for the combination (as determined by the clinical team with prescribing responsibility). A copy of the data collection form can be found in Appendix C. Submission of data Each Trust was allocated a code that was known only to itself and POMH-UK. The Trust s Local Project Teams were asked to allocate codes to participating
wards and eligible patients. The key to these codes was held by the Trust and not known to POMH-UK. Trusts were given the option of using an additional identifier (to code, for example, individual consultants or clinical teams); again, these codes were known only to the Trust team. Data coded in this way were entered onto an internet-based form and submitted to POMH-UK via a secure website. Data cleaning Data were cleaned to correct instances of obvious data entry error. Details of corrections are held on file by POMH-UK; please contact pomhuk@cru.rcpsych.ac.uk if you wish to examine these. Data analysis Definition of high-dose A high-dose is defined as a prescribed total daily dose of a single antipsychotic which exceeds the upper limit stated in the British National Formulary (BNF), or a prescribed total daily dose of two or more antipsychotics which exceeds the BNF maximum using the percentage method. The percentage method converts each drug dosage to a percentage of the respective maximum recommended dose; where the percentages added together are greater than 100%, this is a high-dose. Definition of combination A patient was considered to be prescribed an antipsychotic combination if they were currently prescribed two or more different antipsychotics. A prescription for the same antipsychotic drug by two or more different routes of administration (e.g. oral and intramuscular) or conditions for administration (e.g. regular and prn) was not classed as a combination. Data were analysed at 3 levels: 1. National data. This section describes the demographic and clinical characteristics, as well as the prevalence of high-dose and combination prescribing, in the total sample. The data were analysed in a variety of ways to facilitate understanding of the national picture and stimulate discussion. 2. Trust level data. The analyses conducted on the national data were repeated for each Trust individually. This allows teams in each Trust to compare the demographic and clinical characteristics of their patients, and their prescribing practice, with the anonymised data from each of the other participating Trusts and the national data set as a whole. 3. Ward level data. For each ward, the proportion of patients prescribed high-dose and the proportion receiving combined antipsychotics were calculated. This allows Trust teams to compare prescribing practice across each of their participating wards, and with the national data.
Data were analysed using SPSS version 14. All figures are rounded to zero decimal places for clarity of presentation. Therefore the total percentages for some charts or graphs add up to 99% or 101%. The abbreviation TNS on some charts refers to the combined data set of the total national sample. The Local Project Team lead for each participating Trust has been sent an Excel dataset containing their Trust s data exactly as it was submitted to POMH-UK. This allows Trusts to conduct further analyses on their own data should they wish; for example, examination of prescribing practice at the level of individual clinical teams. Change interventions Between March and November 2006 nine change interventions were offered to Trusts to support quality improvement on participating wards (in addition to the baseline reports which illustrated to members their practice benchmarked against other wards and Trusts, the national sample and recognised standards). These change interventions are described briefly below. 1. Benchmarked audit report. Each Trust received an individualised baseline audit report that benchmarked prescribing practice on each of their participating wards against the Trust as a whole, other participating Trusts and the total national sample. Personalised feedback of audit data is known to be a catalyst for change (Jamtvedt et al, 2006, Patrick et al, 2006). 2. Powerpoint slide presentation with speakers notes to assist local presentation of the evidence base, guideline recommendations and feedback of benchmarked audit data. Educational meetings that contain an interactive component are known to change practice (O Brien et al, 2001). Local opinion leaders may also have a positive effect on practice (O Brien et al, 1999). 21 Trusts ordered this intervention. 3. Bringing about change workshop. This intervention was aimed at clinical team leaders and covered the skills required to successfully bring about change (Constable, 2005). Delegates from 10 Trusts attended this training. 4. Academic detailing workshop. This intervention was aimed at hospital pharmacists and covered the skills required to communicate an evidencebased message. Academic detailing is widely used by the pharmaceutical industry and is known to influence practice (O Brien et al, 1997). Participants from 27 Trusts attended this training. 5. Ready reckoner. This chart facilitated calculation of the cumulative dose of combined antipsychotics, increasing clinicians awareness of how combining antipsychotics can lead to high-dose. Dissemination of information may have a small impact on practice (Grimshaw et al, 2004). 28 Trusts ordered this intervention. 6. Time-series chart. This enabled clinical staff to chart their use of combined antipsychotics and high-dose over time so that trends could be identified. 28 Trusts ordered this intervention.
7. Educational workbook based on CBT principles. This encouraged reflective practice around the use of high-dose and combination antipsychotics, particularly the use of as required medication which was the major cause of combined antipsychotics and high-dose in the baseline audit. Nurses requests for more medication to be prescribed has been shown to be a major influence on high-dose prescribing (Ito et al, 2006). Identifying and targeting individual barriers to change may positively influence practice (Shaw et al, 2005). 27 participating Trusts ordered this intervention. 8. Reminder stickers for prescription charts. These facilitated easy identification of prescriptions for combined and/or high-dose antipsychotics. Patient specific reminders of clinical guidelines may influence practice (Jamtvedt et al, 2006). 32 Trusts ordered this intervention. 9. Educational poster. This outlined the recommendations in clinical guidelines that it should be routine clinical practice to use a single antipsychotic in a standard dose. The risks associated with prescribing outside guidelines were also summarised. Dissemination of information may have a small impact on practice (Grimshaw et al, 2004). 20 Trusts ordered this intervention.
Glossary of terms Antipsychotics The drugs referred to in this report as antipsychotics are all those included in sections 4.2.1 and 4.2.2 of the British National Formulary (BNF). Each drug was listed on the data collection form (see Appendix C). First-generation antipsychotics (FGA) First-generation antipsychotics are the older drugs, also known as typicals or typical antipsychotics, as outlined in the BNF, sections 4.2.1 and 4.2.2 (with the exception of risperidone long acting injection). Second-generation antipsychotics (SGA) Second-generation antipsychotics are the newer drugs, also known as atypicals or atypical antipsychotics, as outlined in the BNF, section 4.2.1 and 4.2.2 (risperidone long acting injection only). Antipsychotic Combination Where two or more different antipsychotics are prescribed together this is referred to as combination prescribing. For example, clozapine and olanzapine prescribed together is a combination prescription, but risperidone prescribed for both oral and depot administration is not a combination as this is the same drug administered via two different routes. High-dose of antipsychotic A high-dose is defined as a prescribed total daily dose of a single antipsychotic which exceeds the upper limit stated in the British National Formulary (BNF) or a prescribed total daily dose of two or more antipsychotics which exceeds the BNF maximum using the percentage method. The percentage method converts each drug dosage to a percentage of the respective maximum recommended dose; where the percentages added together are above 100%, this constitutes a highdose. PRN PRN means as required. PRN prescribed doses are included in calculations of total daily dose. NICE guidelines The National Institute for Health and Clinical Excellence (NICE) provides guidance on the promotion of good health and the treatment and prevention of ill health based on current evidence from systematic reviews and randomised controlled trials (RCTs); www.nice.org.uk.