Comments to regional problems of analgesic risk perception.

OBJECTIVES: The aim of our study was to analyse analgesic risk perception and then to compare analgesic drug choice among general practitioners. METHOD: The structured questionnaire was used and completed during continuous medical education lectures. Series of targeted open or close questions and visual analog scale (VAS) to determine drug risk perception were used. Slovak general practitioners attending continuous medical education lectures during 2004-2005 were invited to participate in the study. Group 1 consisted of respodents from Bratislava (capital city of Slovakia, n = 245) and group 2 consisted of general practitioners from 3 other cities (middle and eastern Slovakia, n = 325). Data were compared to reported adverse drug reactions. RESULTS: Quarter of doctors 25.3% (n = 62), (25.2% (n = 82) respectively), considered non-steroidal anti-inflammatory drugs to be the safest group of analgesics. Gastrointestinal damage in general was perceived as most common adverse drug reaction. 72.41% (75.94% respectively) of respondents considered analgesics as exactly or probably danger. Perceived drug risk labeled on VAS was 4.23 (SD 1.52), (3.22 (SD 2.19) respectively) (p < 0.05). Total number of reported adverse drug reactions in years 1998-2002 was 3249, 412 were related to analgesic use. Specific organotoxic adverse drug reactions (nephrotoxicity, etc.) were reported rarely. CONCLUSION: The actual perception of analgesic risk in Slovakia seems to be generally inadequate. We found only a low support of spontaneous adverse drug reactions reporting to the national monitoring system (Tab. 1, Fig. 2, Ref. 11).

Safety of nateglinide as used in general practice in England: results of a prescription-event monitoring study.

Nateglinide (Starlix((R))) is licensed for the treatment of Type 2 diabetes in patients inadequately controlled with metformin. The study objective was to monitor the safety and use of nateglinide prescribed by primary care physicians (GPs) in England, using the observational cohort technique, Prescription-Event Monitoring. Exposure data were derived from dispensed nateglinide prescriptions issued October 2001-June 2004; demographic and outcome data, from questionnaires sent to patients' GPs at least 6 months after patients' first prescription. Incidence densities (IDs; number of first reports of an event/1,000 patient-months exposure) were calculated for month 1 (ID(1)), months 2-6 (ID(2-6)); rate differences [ID(1)-ID(2-6) (+99% CI)] were examined. Cohort comprised 4,557 patients, median age 60 (IQR 51, 68 years); 2,439 (53.5%) male; 3,463 (76.0%) received nateglinide in combination with metformin. GPs reported 1,625 reasons for stopping in 1,474 (32.3%) patients and 80 events as adverse drug reactions in 66 (1.5%) patients. Events associated with starting treatment included nausea/vomiting [ID(1)-ID(2-6) 9.6 (99% CI 5.3, 13.9)], malaise/lassitude [ID(1)-ID(2-6) 6.03 (99% CI 2.2, 9.9)]. No serious hypersensitivity reactions were reported. Two pregnancies (< 0.1%) and 73 deaths (1.6%) were reported. Nateglinide appeared to be generally well tolerated when used in combination with metformin for the treatment of Type 2 diabetes.

Safety profile of orlistat: results of a prescription-event monitoring study.

INTRODUCTION: Orlistat, the first of a new class of drugs for the treatment of obesity, was launched in the UK in December 1998. The prescribing information recommends that treatment with orlistat should be discontinued after 12 weeks if the patient has not achieved a specified loss of weight. OBJECTIVE: To monitor the safety of orlistat prescribed in the primary care setting in England using prescription-event monitoring (PEM). METHODS: A postmarketing surveillance study using the observational cohort technique of PEM. Patients were identified from dispensed prescriptions issued by primary care physicians for orlistat between December 1998 and November 1999. The outcome data were event reports obtained by sending questionnaires (green forms) to the prescribing doctor at least 6 months after the first prescription for an individual patient. Incidence densities, expressed as number of first reports of an event/1000 patient-months of exposure, were calculated. Significant differences between incidence densities (IDs) for events reported in the 1st month (ID(1)) and months 2 and 3 (ID(2-3)) of exposure were regarded as potential signals. Reasons for stopping orlistat were analysed. Follow-up information was requested for selected events and used to assess the causal association with orlistat. RESULTS: Green forms containing clinically useful information on 16 021 patients (median age 45 years (interquartile range 35-54); 80.1% females) were received. The events reported most frequently during the 1st month of treatment were 'not effective' (639; 4.0% of cohort), diarrhoea (371; 2.3%) and weight loss (230; 1.4%). Twelve clinical adverse events were identified for which ID(1) was significantly greater than ID(2-3). These included non-specific events (e.g. intolerance, malaise/lassitude, unspecified side effects), weight loss and vaginitis/vulvitis. The remaining events were gastrointestinal in nature and included diarrhoea, pain abdomen, flatulence, nausea/vomiting, rectal discharge, faecal incontinence and 'gastrointestinal unspecified' events. A similar pattern of predominately gastrointestinal events was seen for reasons for stopping and suspected adverse drug reactions. Review of selected events for causality revealed 45 events which were assessed as possibly or probably related to orlistat. CONCLUSIONS: This study shows that orlistat is fairly well tolerated. The safety profile of orlistat was similar to the prescribing information and experience reported in the literature.

Observational cohort study to monitor the use and safety of carvedilol in the treatment of heart failure in clinical practice in England--1st interim report.

AIMS: This is an interim report of a prospective observational cohort study to monitor the safety and tolerability of carvedilol in clinical practice when prescribed for heart failure in England. The utilization of carvedilol, management of adverse events in the community and the symptomatic progression of heart failure were examined. It is a non-interventional, observational cohort surveillance study using questionnaires sent to the patients' general practitioners. METHODS: The general practitioners (GPs) of the patients identified by the Prescription Pricing Authority were sent an eligibility questionnaire if the patient had been newly prescribed carvedilol between September 1999 and July 2001. Further questionnaires requesting baseline clinical information about eligible patients (carvedilol indicated for cardiac failure) and subsequent event data were sent to consenting GPs at 12 months. The questionnaires also requested specific information about initiation and supervision of treatment, including severity of heart failure and treatment withdrawal. The data were analyzed by using descriptive statistics. In addition, the incidence densities (ID) of each event (per 1000 person-months of treatment) were calculated, ranked and the difference between the ID of each event in the first (ID1) and subsequent 5 months of exposure (ID2) was tested by constructing 99% confidence intervals (CI). Selected events of clinical interest would be followed-up for further information to enable causal association with carvedilol to be assessed. RESULTS: In this interim report we have data on a cohort of 847 eligible patients for whom we have received information from the first follow-up questionnaire. The treatment of carvedilol was initiated in a majority of cases by hospital specialists (87%, n = 735), however, for most of them, further supervision of treatment was done under shared care between GPs and hospitals (70%, n = 595). More than 90% of the patients were started on carvedilol in the recommended dose range for the management of cardiac failure. Amongst the patients where NYHA grade of cardiac failure was expressed, the majority of patients treated with carvedilol had NYHA II (37%, n = 281) and III (40%, n = 297) symptoms at the start of carvedilol treatment. On treatment with carvedilol, improvement in NYHA status was reported for 43% (n = 364) of this cohort, whilst less than 2.5% (n = 20) of the patients deteriorated. The events reported with the highest ID1 were malaise/lassitude (14.6), dizziness (12.2), cardiac failure (9.7), dyspnea (9.7) and hypotension (8.5). The most common events reported as reasons for stopping carvedilol were "drug not effective", "dyspnea", "dizziness" and "lassitude". No events were identified as new signals (according to the ID1-ID2 statistic) of adverse events associated with carvedilol. There have been no events identified in this cohort that have required specific follow-up at the time of writing this paper. CONCLUSIONS: In summary, the interim results show that there is effective cooperation between hospital specialists and GPs in the use of carvedilol in the management of patients with heart failure. It also shows that carvedilol was well-tolerated and that patients with mild and moderate heart failure benefit symptomatically when treated with carvedilol in routine clinical practice.

Evaluation of the safety of sildenafil for male erectile dysfunction: experience gained in general practice use in England in 1999.

OBJECTIVE: To examine the safety of sildenafil, the first of the phosphodiesterase type 5 inhibitors licensed for the treatment of male erectile dysfunction (ED), as used in general medical practice in England, quantifying the incidence of a range of events in patients treated with sildenafil, and identifying any previously unrecognized adverse drug reactions. METHODS: In a postmarketing observational cohort study using prescription-event monitoring (PEM), exposure data were derived from dispensed prescription details for patients who started treatment between April and August 1999. Outcome data were derived from "green form" questionnaires (GFs) returned by general practitioners (GPs). RESULTS In all, 24 835 (54.7%) of GFs posted to GPs were returned, of which 22 473 contained useful data for 22 471 male and two female patients. The major primary indications/clinical context of prescribing were impotence (16 583, 73.8%) and diabetes mellitus (183, 0.8%); 145 events were reported as adverse drug reactions to sildenafil. GPs recorded 3951 reasons for stopping sildenafil, and ischaemic heart disease (IHD) in 135 patients was the commonest clinical reason reported. The clinical condition reported most frequently in the first month of observation was diabetes mellitus and/or hyperglycaemia (in 99 events). A standardized mortality ratio (SMR) for deaths caused by IHD in the first 8893 of 22 473 patients was 31.41 (95% confidence interval 18.29-50.29), using the comparator population of males in England in 1998. CONCLUSION: This study identified the safety profile of sildenafil as used in the community, showing no unexpected events. The SMR analysis of deaths from IHD provided no evidence to suggest a higher incidence of deaths in the study cohort than in the male population in England.

Prospective observational cohort safety study to monitor the introduction of a non-CFC formulation of salbutamol with HFA134a in England.

OBJECTIVE: To monitor the safety of a salbutamol MDI with a hydrofluoroalkane propellant (Ventolin Evohaler) during its introduction into primary care use in England. METHODS: Prospective observational cohort study. 1,365 GPs in England submitted data on 10,472 regular users of Ventolin MDI, over five 3-month periods of observation between October 1, 1998 and December 31, 1999. The primary aim was to compare event rates occurring before and after the introduction of Ventolin Evohaler. The secondary aim was a comparison of event rates between users of Ventolin Evohaler and Ventolin MDI. The main outcome measures were: indication for use of Ventolin MDI, assessment of disease severity, event rates during each period of observation; deaths, pregnancies, reported adverse drug reactions and reasons for discontinuation of MDI. Event rates were adjusted using a ratio for under-reporting derived from a validation study on 4.6% of the study population and stratified by severity of indication. RESULTS: The primary indication was asthma in 94%, distributed by severity as 47% mild, 44% moderate and 9% severe; 13% were children. By October 1999, 52.7% of the 8,973 remaining patients had transitioned to Ventolin Evohaler. There was no increase in major or minor events observed following the introduction of Ventolin Evohaler. No serious adverse events, abnormal pregnancy outcomes or deaths have been related to Ventolin MDI or Ventolin Evohaler. The validation study showed a degree of under-reporting. CONCLUSION: These results on a large cohort of community patients in England indicate that Ventolin Evohaler is well tolerated among asthmatics.

The safety of valsartan: results of a postmarketing surveillance study on 12881 patients in England.

Valsartan is a second class of angiotensin II receptor antagonist, indicated for the treatment of hypertension. The objective of the study was to monitor the safety of valsartan using the technique of prescription event monitoring (PEM), in patients who were prescribed this drug by general practitioners (GPs) in England. PEM is a noninterventional observation cohort technique. Exposure data were obtained from dispensed prescriptions issued between December 1996 and November 1998. Outcome data were obtained by sending questionnaires to prescribing GPs. The cohort comprised 12881 patients. Events most frequently reported as suspected adverse drug reactions were malaise/lassitude (37; 0.3% of total cohort), dizziness (19; 0.1%), and unspecified side effects (57; 0.4%). Events with the highest incidence density (ID(1) per 1000 patient-months of treatment) in the first month of treatment were malaise/lassitude (15.6), dizziness (11.8), and headache/migraine (10.9). Most frequent reasons for stopping valsartan were not effective (847; 6.6% of total cohort), malaise/lassitude (265; 21%), and dizziness (146; 1.1%). No unexpected serious adverse events were identified. Other events assessed as possibly related to valsartan use were impotence (37), dizziness (19), cough (9), facial oedema (5), hyperkalaemia (3), and angioneurotic oedema (1). There were four reports of exposure during pregnancy and 203 deaths (1.5%) in this cohort. In conclusion, this study monitored the safety profile of valsartan in a large cohort of patients in general practice in England. No untoward features other than dizziness were identified that were not mentioned in the prescribing guidance.

A post-marketing observational study to assess the safety of mibefradil in the community in England.

OBJECTIVES: To conduct a post-marketing observational cohort study to assess the safety of mibefradil in the community, using Prescription-Event Monitoring (PEM). METHOD: Data were collected and analyzed on patients prescribed mibefradil by 1,996 General Practitioners (GPs) throughout England. Incidence densities were calculated for all reported events and selected events were followed-up by means of further questionnaires. RESULTS: The study was terminated early due to the voluntary withdrawal of mibefradil from the market because of potential drug interactions. A cohort of 3,085 patients was recruited, with a mean age of 64.5 years. The major indication for use was hypertension (55% of the cohort), the indication was not specified in 33% of patients. 80% of GPs expressing an opinion rated mibefradil as effective. The major reason for stopping was withdrawal from the market (2,342 patients). The commonest reported adverse events and reasons for stopping were malaise/lassitude, dizziness, edema and headache. Seven clinically serious reports of bradycardia/collapse were considered to be possible adverse drug reactions (ADRs) to mibefradil. All were in the elderly (> 65 years), 6 were considered to be a result of possible drug interactions. In total, 11 possible drug interactions occurred. Nine (8 reports of bradycardia and 1 of syncope) involved beta-blockers. Another, a report of collapse and severe bradycardia, occurred in a patient who had started a dihydropyridine calcium channel blocker within 24 hours of stopping mibefradil and the other was a report of palpitations and dyspnea in a patient on concomitant digoxin and sotalol. None of the 53 deaths occurring during the study was attributed to mibefradil. CONCLUSION: Mibefradil was only available on the UK market for 6 months before it was withdrawn from the market because of potential drug interactions. With respect to the reasons leading to its withdrawal, in this cohort of 3,085 patients, 11 possible drug interactions were detected (6 clinically significant) involving beta-blockers, a dihydropyridine calcium channel blocker and digoxin and/or sotalol. PEM can contribute to the understanding of ADRs caused by drug interactions occurring in real-life settings.

Prescription-event monitoring and reporting of adverse drug reactions.

Newly marketed drugs in the UK are marked with a black triangle, indicating that doctors should report all adverse drug reactions associated with them to the Committee on Safety of Medicines (CSM). However, under-reporting of adverse reactions is frequent. Our aim was to establish what types of adverse reactions are under-reported to the CSM by family doctors who work in England. We used prescription-event monitoring data obtained for 15 newly marketed drugs. Only 9% (376) of 4211 events found on prescription-event monitoring were reported to the CSM. However, 53% (27) of 51 events classified as serious adverse drug reactions were reported. Overall, serious events were five times more likely to be reported to the CSM than non-serious events. Our results should not be extrapolated to calculate incidence rates of adverse drug reactions in the community from spontaneous reports.

Comparative study of mortality rates and cardiac dysrhythmias in post-marketing surveillance studies of sertindole and two other atypical antipsychotic drugs, risperidone and olanzapine.

Sertindole (Serdolect), an atypical antipsychotic, was voluntarily suspended in the European Union in 1998 following regulatory concerns over reports of serious cardiac dysrhythmias and sudden unexpected deaths. The reported causes of death, their frequency, prolongation of the rate corrected QT interval (QTc) and cardiac dysrhythmias in patients prescribed sertindole were compared with those for patients treated with two other atypical antipsychotics. All patients in England, prescribed atypical antipsychotics by general practitioners during each drug's immediate post-marketing period, were identified using an observational cohort technique, prescription-event monitoring. Mortality rates in the sertindole cohort were compared with those in a comparator cohort using standardized mortality ratios and incidence rate ratios. Cardiovascular events were reviewed and followed up to identify cases of prolongation of QTc interval. There was no statistically significant difference in mortality rates between sertindole and the comparator cohort, although confidence intervals (CI) were wide due to small numbers in the sertindole cohort. A much smaller number of patients were prescribed sertindole than the other antipsychotics. Six cases of prolongation of QTc interval were identified in 462 patients (1.3%, 95% CI 0.5-2.8) treated with sertindole and one with unspecified electrocardiogram changes in the comparator cohort of 16,542 patients. This study contributes to the understanding of the occurrence of prolongation of QTc interval during clinical use of sertindole, the incidence of which was similar to that in clinical trials. Although no statistically significant difference was shown in mortality rates between sertindole and comparator cohort, the sertindole cohort was too small to rule out an association between the use of this drug and cardiovascular deaths.

Troglitazone and liver function abnormalities: lessons from a prescription event monitoring study and spontaneous reporting.

OBJECTIVES: To investigate whether there were any cases of liver function abnormalities possibly associated with troglitazone use in general practice in England. DESIGN: A prescription-event monitoring (PEM) study was undertaken between October 1997 and December 1997. SETTING: Data from prescriptions were obtained electronically for the troglitazone cohort in the immediate postmarketing period. STUDY PARTICIPANTS: Event data were obtained for a total of 1344 patients. RESULTS: Troglitazone was effective in 394 (75%) of the 529 patients for whom an opinion was given. The most frequent reasons for stopping treatment related to drug tolerability were malaise/lassitude (16 reports), abnormal liver function tests (II reports) and nausea/vomiting (9 reports). The major cause of stopping troglitazone was because the drug was withdrawn from the market (1101 reports). 30 patients with liver dysfunction were identified from the cohort. In 9 of these patients there were alternative explanations for the liver dysfunction and hence these patients were not followed up further. 21 patients were followed up, for whom 19 questionnaires were returned. In 5 patients their liver dysfunction was assessed as possibly related to troglitazone, in 6 patients the liver dysfunction was unlikely to be attributed to troglitazone, while in 7 patients it was difficult to assess the causality because of limited information and confounding factors. The remaining patient was not included as this individual did not fit the inclusion criteria of the study. CONCLUSION: Although the cohort is small (the drug was available for only 3 months in the UK), 5 patients with abnormal liver function, considered possibly related to troglitazone were detected in this PEM study. It is possible for PEM to contribute to the elucidation of safety signals in the UK.

The pharmacovigilance of olanzapine: results of a post-marketing surveillance study on 8858 patients in England.

Olanzapine is an 'atypical' antipsychotic indicated for the treatment of schizophrenia. We analysed adverse events (AEs) reported in primary practice in England. Dispensed prescriptions issued between December 1996 and May 1998 provided exposure data. Questionnaires sent to general practitioners provided outcomes. Frequently reported AEs were: drowsiness/sedation (n = 19), extrapyramidal disorder (n = 13) and unspecified side-effects (n = 33). Events with highest incidence density in first month and reason for stopping were: drowsiness/sedation [n = 153, incidence density (ID)1 18.9], weight gain (n = 117, ID1 8.9) and malaise/lassitude (n = 65, ID1 5.2). Extrapyramidal disorders were more common in elderly population (> 70 years, ID1 3.6, risk 26.0 per 1,000 patients) compared to < 70 years (ID1 1.1, risk 8.4 per 1,000 patients). Serious suspected adverse reactions were neuroleptic malignant syndrome (n = 1) and angioneurotic ooedema (n = 2). There were eight reports of diabetes mellitus assessed as possibly due to olanzapine. Diabetes mellitus was an unlabelled AE and possible signal generated by prescription-event monitoring.

Profiles of hepatic and dysrhythmic cardiovascular events following use of fluoroquinolone antibacterials: experience from large cohorts from the Drug Safety Research Unit Prescription-Event Monitoring database.

OBJECTIVE: To investigate how frequently serious dysrhythmic cardiovascular, and hepatotoxic events are reported during routine clinical use of fluoroquinolones (quinolones) in general practice. DESIGN: Cohorts prescribed quinolones (cohort sizes: ciprofloxacin 11 477; enoxacin 2790; ofloxacin 11 033 and norfloxacin 11 110; mean age in each cohort of 48.6 to 57.0 years) were selected from the Drug Safety Research Unit's Prescription-Event Monitoring (PEM) database. The monitoring periods were November 1988 to January 1989 for ciprofloxacin; April 1989 to January 1991 for enoxacin; May 1991 to December 1991 for ofloxacin and October 1990 to October 1991 for norfloxacin. Data collected over the total PEM surveillance period on selected gastrointestinal events were extracted and reviewed to identify possible hepatic events, together with selected cardiovascular events associated with dysrhythmias. For each quinolone, times to onset of the event and patient-months of observation (denominator values) were calculated. The analysis was based on two observation periods: rate of event during the first 7 days following dispensing of a prescription for each drug (W(1)), and rate of event during the second to sixth week inclusive (W(2)). RESULTS: Scrutiny of original green forms revealed no evidence of drug-induced hepatic dysfunction within 42 days of drug administration for any of the quinolones monitored. No evidence was found of drug-induced dysrhythmic events associated with enoxacin within 42 days of drug administration. Of the other quinolones, 'atrial fibrillation' was reported most often within 42 days following ciprofloxacin administration, with no change in event rate over that time, crude relative risk (CRR)[W(1)/W(2)] 1.0 [95% confidence interval (CI) 0.02 to 8.92]. Other less serious events associated with dysrhythmia were reported with varying incidence within 42 days of quinolone administration. The crude rate of palpitation did not change significantly over that time for ciprofloxacin, ofloxacin and norfloxacin: CRR 0.83 (95% CI 0.02 to 6.86), 2.00 (95% CI 0.19 to 12.20) and 4.99 (95% CI 0.06 to 391.94), respectively. Syncope and tachycardia were also reported for ofloxacin [CRR 9.99 (95% CI 0.52 to 589.49 for both events)] and ciprofloxacin [1.0 (95% CI 0.02, 8.92)] and 2.50 (95% CI 0.04, 47.96) for syncope and tachycardia, respectively]. CONCLUSION: It cannot be ruled out that some rare hepatic and dysrhythmic events associated with quinolones may be drug related. The primary purpose of PEM is signal generation. Compared with the other quinolones, ciprofloxacin was associated with the highest number of reports of dysrhythmic cardiovascular events occurring within 42 days of administration. This requires further investigation by other types of epidemiological study.

Montelukast and improvement of eczema: observations from a prescription event monitoring study in England.

OBJECTIVE: Montelukast is an orally administered cysteinyl receptor antagonist, approved for the treatment of asthma. There is pharmacological plausibility of its effectiveness in the treatment of other immunologically mediated conditions such as eczema and urticaria. The objective of this study was to determine whether there are any beneficial effects of montelukast on eczema and urticaria. METHOD: A non-interventional observational cohort study was conducted between February 1998 and December 1998 using Prescription-Event Monitoring (PEM). During PEM studies, patients are systematically identified from dispensed prescription data and questionnaires are sent to the prescribing general practitioner (GP) asking them to report events occurring during and after treatment. In this study, events reported as eczema or urticaria improved were identified. A simple questionnaire was sent to the GPs for additional information. RESULTS: The cohort comprised 15,612 patients, in which 16 reports of eczema or urticaria improved were identified. Questionnaires were sent to the GPs for additional information. Fifteen of the 16 questionnaires were returned. In 5 cases the GPs thought that there was an improvement of eczema or urticaria with montelukast treatment in patients who had history of longstanding eczema or urticaria. Of the remaining 11 cases there was an alternative explanation for the improvement of eczema or urticaria in 10 cases and one was unassessable. CONCLUSION: PEM is conducted to monitor the safety of medicines, and doctors report events including improvement in pre-existing conditions. Although the number of cases of improvement of eczema or urticaria in this cohort is small, there is a possibility that leukotriene inhibitors may be helpful in the treatment of these diseases. Further studies are needed to provide evidence as to whether montelukast will have a role in the treatment of these conditions.

Interim report on the incidence of visual field defects in patients on long term vigabatrin therapy.

The incidence was determined of visual field defects in patients in the Prescription-Event Monitoring (PEM) study, treated for > or =6 months with the antiepileptic drug vigabatrin. Questionnaires were sent to the general practitioners of the 7228 patients in the PEM study, treated with vigabatrin for > or =6 months, to ascertain whether any serious adverse events, including changes in vision, had occurred. Ophthalmologists were asked to give the results of the perimetry tests and their opinion of the cause of the visual field defect for those patients referred to them. 6793 (94%) of the 7228 questionnaires were returned. 5090 (75%) contained clinical data. 328 patients were deceased. Seventy seven cases identified from the 4762 surviving patients, are being followed up with ophthalmologists. To date, 2 weeks after posting, 12 cases of visual field defect have been confirmed by formal perimetry tests. Ten of these 12 were considered to be probably or possibly related to vigabatrin use, giving an incidence of 2.0 per 1000 patients compared with 0.4 per 1000 patients in the original PEM cohort. These interim results show a substantial increase in the incidence of visual field defects associated with long term use of vigabatrin and demonstrate that PEM can be used to assess long latency adverse events.

The use of newly marketed drugs in children and adolescents prescribed in general practice.

This study determined the licensed and 'off label' (outside the terms of the licence) use of newly marketed medicines in children (2-11 years) and adolescents (12-17 years), by general practitioners in England. In addition, the incidence rates during the first month of therapy (ID(1)) for three adverse events, in these groups were compared with those of adults (> or =18 years). The use of these drugs was monitored in 63 individual prescription-event monitoring (PEM) studies, conducted to monitor the safety of these medicines. Patients and drug exposures were identified from dispensed prescriptions. Outcome data (events and demographic information) were obtained from questionnaires. Although only six of these 63 drugs were licensed for use in children, 44 of the 63 drugs were used to treat children. For the majority of the drugs there was no specific reference to adolescents in the data sheets therefore it has been assumed that the drugs were licensed for those aged > or =12 years unless specified otherwise; 55 have been taken as licensed for use in adolescents. Over 690,000 patients were included in the 63 PEM studies, 9081 (1.3%) of these were children and 15,256 (2.2%) were adolescents. 78% of the 9081 children and 93% of the 15,256 adolescents were treated with 'licensed' drugs. There was a significant difference in the incidence rate for rash and nausea/vomiting, two adverse events commonly reported during treatment with lamotrigine, between children and adolescents compared to adults. This survey has shown that although only a small proportion (10%) of newly marketed drugs were licensed for use in children the majority of children (78%) were treated with these licensed products but 22% of children received drugs 'off label' during the first few years that the drug was marketed and a small number of children and adolescents were given drugs contraindicated in these age ranges.

Prevalence of Churg-Strauss syndrome, vasculitis, eosinophilia and associated conditions: retrospective analysis of 58 prescription-event monitoring cohort studies.

PURPOSE: Churg-Strauss syndrome is characterised by hypereosinophilia, systemic vasculitis and asthma. The cause is usually unknown, but there have been reports of an association with particular drugs, including anti-asthma drugs. Our aim was to estimate the prevalence of Churg-Strauss syndrome and related conditions in post-marketing safety studies of new anti-asthma drugs. METHODS: We accessed the prescription-event monitoring (PEM) database of the Drug Safety Research Unit (DSRU). This database currently has information on 622 294 patients observed during 58 completed PEM studies of individual drugs, including 35 799 patients in studies of new anti-asthma drugs. RESULTS: Overall, four cases of Churg-Strauss syndrome were identified during these studies, giving a period prevalence rate of 6.8 (95% confidence limits [CL]: 1.8-17.3) per million patient-years of observation. The period prevalence of Churg-Strauss syndrome was significantly greater in the PEM cohorts of anti-asthma drugs (64.4 million patient-years of observation; 95% CL: 13.3-188.1), than the other PEM drug cohorts (1.8 per million patient-years of observation; 95% CL: 0.05-10.2) (rate ratio: 35.1; 95% CL: 2.8-1839.9; p=0.002). CONCLUSION: Our data provide estimates of the prevalence of Churg-Strauss syndrome and related conditions in cohorts of asthmatic patients, which may be useful during the pharmacovigilance of new asthma drugs, including the new leukotriene receptor antagonists.