Specific Studies Using Intensive Medicines Monitoring Programme Data

SPECIFIC STUDIES USING INTENSIVE MEDICINES MONITORING PROGRAMME DATA

In addition to the routine data analysis performed for every monitored medicine, the cohort and event databases of the IMMP are increasingly being used for specific pharmacoepidemiology studies. Some examples of studies that have been performed are discussed in this section.

STUDIES OF INTRA-UTERINE DEVICES

The IMMP has performed unique post-marketing safety studies on IUDs and has effectively adapted PEM methodology for this purpose (Zhou, Harrison-Woolrych and Coulter, 2003). Both the copper IUD Multiload Cu375 and the levonorgestrel-releasing device Mirena have been monitored by the IMMP. Cohorts were established with the use of registration forms supplied (by the manufacturer) with each IUD. Doctors completed the registration form at the time of IUD insertion and returned it to the IMMP. Follow-up questionnaires were then sent annually for each woman, usually to her GP, but often to family planning doctors or gynaecologists if this was more appropriate.

For Multiload Cu375, in addition to the routine IMMP reporting (Coulter, 1997), two specific studies were conducted using a cohort of over 16 000 women who used this device in NZ during a 10-year period. The first study was an analysis of insertion problems and reported an overall incidence of approximately 2% for failed/difficult insertion and an incidence of about 1% for adverse reactions to insertion (Harrison-Woolrych, Ashton and Coulter, 2002). This study, which is thought to be the largest study of IUD inser-tion published to date, also identified nulliparity and experience of the inserting doctor as risk factors for inserting problems.

The second Multiload Cu375 study investigated the incidence of uterine perforation with this device (Harrison-Woolrych, Ashton and Coulter, 2003a). The rate of 1.6 per 1000 insertions was higher than previ-ously reported, and one reason for this was thought to be the long period of intensive follow-up (10 years) in the IMMP study compared with other studies. Most uterine perforations (86%) were not diagnosed at the time of insertion with some remaining undiagnosed for several years.

The IMMP also performed a comparative study of the Multiload Cu 375 device and the levonorgestrel IUD Mirena (Harrison-Woolrych, Ashton and Coulter, 2003b). This reported a significantly higher incidence of insertion problems with the Mirena device than with the copper IUD, although difficult insertions were reported in fewer than 4% of Mirena insertions. The levonorgestrel IUD is now widely used in NZ and many other countries for contraception (and menorrhagia) and comparative studies of this kind are of great value in assisting women and their doctors to make appropriate choices (Harrison-Woolrych, 2003).

STUDIES IN SPECIAL POPULATIONS

Using similar methodology to the UK PEM scheme, the IMMP routinely collects information on pregnan-cies in women taking the monitored medicines. All pregnancies are followed up to determine outcome for the mother and baby. In the IUD studies, infor-mation was collected on both inadvertent pregnancies and planned pregnancies after device removal.

Paediatric use of atypical antipsychotic medicines has been a recent focus of interest for the IMMP. A population of children (defined as age 15 years or under) has been identified from the IMMP cohorts, and specific follow-up questionnaires have been sent to child/adolescent psychiatrists and/or the children’s GPs. This study is seeking information primarily on adverse events in children but also on indication for use, the duration of treatment and reasons for stop-ping the medication. Data on the safety of medicines in children are often very limited, and frequently pre-licensing trials do not include a paediatric popula-tion, so post-marketing studies of this type are very important.

DATA-LINKAGE STUDIES

Earlier in this chapter, we explained how the NHI number of every patient in the IMMP cohorts can be linked to national morbidity and mortality databases to perform record linkage studies. Generally, this type of pharmacoepidemiology study is most useful for study-ing deaths and more serious events, which result in hospital admission (GP consultations are not recorded in the NZHIS databases), and NZ is an ideal country in which to perform such studies.

The IMMP is currently developing a data-linkage approach to perform studies on the four atypical antipsychotic medicines  clozapine, olanzapine, quetiapine and risperidone. Having these four cohorts established for the same time gives the potential to perform comparative studies. Routine IMMP follow-up of patients on these medicines (via their doctor) has proved very difficult, and response rates have been significantly lower than for other medicines. Using record-linkage methodology allows a higher propor-tion of patients to be followed up, often in a more time-efficient manner. Information obtained on clin-ical events from national databases can subsequently be checked or supplemented with further detail from patients’ doctors.

There are limitations to data-linkage studies  e.g. usually less detailed information will be obtained than from IMMP follow-up questionnaires, information on risk factors for events is frequently not available and only events resulting in hospital admission or death are recorded. However, for the investigation of specific events, e.g. ectopic pregnancy in Mirena users or sudden cardiac death in patients taking sibu-tramine, data-linkage methodology may be extremely useful.

PHARMACOGENETIC STUDIES

The ability to identify individuals who are suscepti-ble to ADRs has the potential to reduce the personal and population costs of drug-related morbidity. Data from the IMMP cohorts and events databases have been used to initiate pharmacogenetic studies, and it is hoped that this will be another area of future development.

A pilot study has been conducted to investi-gate the methodology of linking PEM studies with pharmacogenetics in the IMMP (Clark et al., 2004). This study used a nested case–control design to investigate whether patients with genetic variants in P-glycoprotein and CYP2C9 are more susceptible to psychiatric or visual disturbances following COX II inhibitor use than matched controls taking the same medicine without experiencing an adverse event. This paper by Clark et al. also discusses some future direc-tions for linking pharmacoepidemiology studies with pharmacogenetic investigations.

COMMUNICATING RESULTS OF INTENSIVE MEDICINES MONITORING PROGRAMME STUDIES

A high priority for the IMMP is provision and feed-back of information to prescribers and patients. Every spontaneous report submitted to the IMMP receives a detailed individual reply that includes (anonymized) information from the IMMP databases. Similarly, information is provided in response to telephone or other enquiries on a daily basis. In addition, summaries of IMMP work on particular medicines are regularly provided to individual doctors or those working in a specialist area – e.g. a summary of work on the Mirena IUD was e-mailed to all family plan-ning clinics in NZ via the Family Planning Associa-tion network.

Reports of IMMP data and studies are regularly provided to the Ministry of Health (Medsafe) and the MARC, with summaries of these reports published in the MARC minutes. IMMP reports may raise new signals or other issues for consideration or might provide reassuring data on emerging issues. The IMMP also provides reports for information to the NZ licensing committee (MAAC) and – with the agree-ment of the NZ Ministry – will provide information to other national committees or other international regu-latory authorities on request. As NZ is a small country, collaboration in pharmacovigilance with other coun-tries is valuable, and data may be pooled and expertise shared.