Long term follow-up studies of users of nonprescription medicines purchased from community pharmacies: some methodological issues

Abstract

Background: Despite the wider availability of medicines to the general public, little is known about their safety when supplied without prescription. Pilot work has already tested 4 methods of recruiting users of ibuprofen purchased from community pharmacies. This paper describes the piloting of a fifth method (a shortened questionnaire), long term follow-up rates of all methods, consistency of reporting of ibuprofen use, and issues relating to possible comparison groups in pharmacovigilance studies.

Methods: A shortened version of a previously tested recruitment questionnaire was used. Eligible study participants were all users, aged over 17 years, purchasing ibuprofen from a research network of community pharmacies (n = 61) in Grampian, Scotland. Postal questionnaires were sent at I week and 2, 6 and 12 months irrespective of the method of recruitment. The follow-up questionnaires collected information about ibuprofen and other drug usage, symptoms and associated health service utilisation.

Results: The shortened form recruited 67% of people issued with a questionnaire. The overall 12-month follow-up rate was 67%, although there were important differences in the rates by method of recruitment. There was reasonable consistency in the reporting of use or non-use of ibuprofen at different follow-up times. In the 12 months after the index purchase, 17% of participants never used any ibuprofen (non-users) and 28% used it for more than 8 weeks in total (long term users). At 12 months, long term users were significantly more likely than short term users (< or =8 weeks total use) or non-users to have experienced dizziness, skin rash, itchy skin and wheeziness in the previous week.

Conclusions: Our pilot work has confirmed the feasibility of recruiting, and following-up over prolonged periods, users of nonprescription medicines. Evidence of long term use of ibuprofen confirms the need for pharmacovigilance studies of this drug, although further work is required to identify a suitable comparison group in order to inform the interpretation of such investigations.