Ziprasidone

We principally undertook a segmented regression analysis of retrospective time-series analysis to assess the effect of various initiatives in Belgium, Ireland, Scotland and Sweden following the introduction of generic risperidone [51 (link)]. The xtmixed command in Stata (version 12) (StataCorp, College Station, Texas, USA) was used to fit a linear random coefficient model with country-specific intercepts. At the time of introduction of generic risperidone into each country, a random shift in intercepts and slopes was allowed to estimate the effect of the introduction. Data on the number of monthly reimbursed prescriptions within each country’s health service for all patients prescribed at least one atypical antipsychotic drug (N05AH03 to 06, N05AL05, N05AX08, N05AX011 to 13) [52 ] up to 20 months before and up to 20 months after the availability of generic risperidone was included. Clozapine was not included in the analysis as it is generally reserved for patients not responding to other atypical antipsychotic drugs because of its side-effect profile [46 ,53 -55 (link)]. Ziprasidone (N05AE04) was also not included. This was in view of its different classification and limited utilisation in practice in a number of European countries, including Sweden [56 ].
A retrospective observational study of the same population dispensed at least one atypical antipsychotic drug was also undertaken in Austria and one of the regions in Spain (Catalonia) from January 2005 (Austria) or January 2006 (Spain) to the end of 2010 (Austria) and September 2011 (Spain). This was because generic risperidone was already available in Austria and Spain in July 2004 and by January 2006 respectively, but only became available later in the four chosen European countries: Ireland in December 2007, Scotland in April 2008, and Belgium and Sweden in January 2009. A retrospective observational study was also undertaken in one English primary care organisation, Bury Primary Care Trust (PCT), between November 2009 and October 2011. The objective was to assess the influence of a request to psychiatrists to consider oral risperidone as first line treatment in new or other suitable patients, where appropriate, now it was available as a generic.
Finally, retrospective observational studies were undertaken on the utilisation of long-acting risperidone injections versus total risperidone (N05AX08) [52 ], which was available throughout the study period, as well as paliperidone (N05AX13) before and after the availability of generic oral risperidone.
The European countries chosen provide a range of differences in geographical location, population size, different approaches to the financing of health care, and different approaches to the pricing of generics and to enhancing the utilisation of generics versus originators [39 ,40 ], which is in line with recommended guidance [57 (link)].
Only administrative databases were used in each country to assess the utilisation and expenditure patterns of the atypical antipsychotic drugs. This is because the perspective of the study was that of health authorities, and they typically have the greatest knowledge concerning existing and planned initiatives and reforms in their countries. The databases, which are regularly audited, are included in Box 1. Box 1 also contains details of patients included within the national health service of each country. This typically includes 100% or close to 100% of the population unless stated (Ireland), given the principles of equity and solidarity within European healthcare systems. There are also typically limited patient co-payments.
The utilisation of the different atypical antipsychotics was calculated in terms of defined daily dose (DDD), which is defined as ‘the average maintenance dose of a drug when used in its major indication in adults’, as this measure is recognised as the international standard to assess utilisation patterns within and between countries [58 ]. The only exception was Bury PCT, where utilisation was measured in terms of prescription items, which is the typical metric used to assess utilisation patterns in England [59 ]. 2011 DDDs were used in line with international guidance [58 ,60 (link),61 (link)].
Separate retrospective observational studies were conducted in Belgium, Scotland and Sweden, again using an interrupted time-series methodology. The objective was to assess whether the changes in risperidone utilisation patterns after the introduction generic risperidone in these three countries were significant [46 ,48 ,56 ].
Subsequently, risperidone utilisation in Belgium, Ireland, Scotland and Sweden was converted into a percentage of total selected atypical antipsychotic utilisation (DDD basis) before and after the availability of generic risperidone (time 0). The objective was to enable meaningful comparisons between the four countries, factoring in differences in population sizes, time when generic risperidone became available, and differences in their database characteristics (Box 1). Utilisation patterns and calculations were verified with the relevant co-authors to enhance the robustness of the study findings.
The percentage of oral risperidone dispensed as generics was also calculated in Belgium, Ireland, Scotland and Sweden. We would expect to see considerable differences in utilisation rates between countries in view of the different policies in each country regarding encouraging the utilisation of generics versus originators [39 ,40 ,62 -65 ]. However as mentioned, a universally low utilisation of generic risperidone would reflect general stakeholder concerns with generic risperidone.
The percentage reduction in expenditure per DDD for oral generic risperidone versus pre-patent loss originator prices was also calculated in Belgium, Ireland, Scotland and Sweden. We chose to compare relative reductions rather than actual prices for generic risperidone as the price components can vary in each country (for example, there are variations in the extent of VAT and relative wholesaler margins), and this approach also avoids currency conversions, both of which can make cross-country price comparisons difficult, especially during times of economic difficulty. In addition, prices of initial or all generics in an appreciable number of European countries are based on pre-patent loss prices [39 ,40 ,62 ,64 ,65 ], and the time periods for the availability of generic risperidone varied considerably between the countries and regions studied. We also did not factor inflation into the calculations because the trend in most European countries is to reduce prices when pharmaceutical expenditure exceeds target budgets [39 ,66 (link)] and, as mentioned, prices of generics in a number of European countries are based on pre-patent loss prices. This is in line with previous studies [39 ,40 ,64 ]. We would again expect to find considerable differences in the prices of generic risperidone between countries, because of the different pricing initiatives and differences in the attractiveness in the generic market [1 ,38 ,39 ,62 ,64 ,67 (link)].
Finally, we calculated the influence of the availability of generic risperidone on subsequent atypical antipsychotic expenditure where possible.
No ethics approval was needed or obtained because only aggregated drug utilisation data was used, without access to specific patient data.

In this preliminary study, patients were randomized to one of two treatments. Patients in the dose reduction group (n = 10) received their current FGAs, but their doses were slowly reduced to a target of 5 mg of HE/day (FGA-DR-condition). Patients in the switch group (n = 13) received ziprasidone at a dose of 160 mg/day, equivalent to 5 mg of haloperidol, given as 80 mg b.i.d. (ZIPRA-condition). This study was part of a clinical research project investigating FGA dose reduction or switching to ziprasidone in long-stay patients with severe chronic schizophrenia treated with high doses of FGAs. The original study was a double-blind randomized study with a 1-year follow-up, with negative symptoms as the primary outcome. The present study focused on neurocognitive function. Study methods have previously been described in detail [34 (link)]. Figure 1 shows a timetable of this study.

The participants of this clinical research project were patients who had severe disabling symptoms, such as aggressive behavior and/or deficit states. Most patients were involuntarily hospitalized in long-stay clinics. These patients had been prescribed various high-dose FGAs for many years. The overall aim of this study was to evaluate whether dose reduction or treatment with ziprasidone would improve clinical outcomes. None of the patients were using clozapine at baseline or had used clozapine in the past, even though they experienced refractory positive and negative symptoms for years. All participants had a DSM-IV diagnosis of schizophrenia or schizoaffective disorder based on a Structured Clinical Interview for DSM disorders (SCID). According to the inclusion criteria, they were older than 18 years, had persistent symptoms for at least 2 years, were on stable medication (no change 3 months prior to inclusion in the study), and were able to comply with the study protocol. All patients gave their consent to use oral capsules of medication or placebo. None had diseases that could pose a medical risk, such as a QTc interval of more than 500 ms, which increases the risk of ziprasidone-associated arrhythmias.
Throughout the study, study medication was given under double-blind conditions. The medication/placebo capsules were prepared by the Central Hospital Pharmacy in The Hague. They were identical in appearance, taste, and smell.

This study had three phases (Figure 1). At baseline, patients were assessed before their medications were changed. In the next phase (the dose-adjustment or switch phase), their medications were changed. In the FGA-DR condition, the dose was gradually reduced to 5 mg of HE/day based on each patient’s starting dose, but was maximally reduced over 24 weeks for baseline FGA doses higher than 30 mg of HE/day (according to a dose reduction table for each individual patient based on the starting dose). It was expected that this slow dose reduction scheme would prevent withdrawal psychosis. In the ZIPRA condition, the current FGA treatment was changed in two steps: 2.5 mg of HE was replaced with ziprasidone at a dose of 80 mg/day, and after 6 weeks, a further 2.5 mg of HE was replaced with ziprasidone at a dose of 80 mg/day (total daily dose of 80 mg b.i.d.). The remaining FGA dose was gradually withdrawn (until 0 mg was reached) as described for the FGA-DR condition. The final treatment and observation phase lasted 1 year.
Patients continued to take non-study medication used at baseline that remained unchanged throughout the study, with the exception of anticholinergic drugs, which were gradually withdrawn during the dose-adjustment and switch phase, but they were reintroduced if the patient experienced extrapyramidal side effects.