Frequency, nature and determinants of pharmacy compounded medicines in Dutch community pharmacies.

Aims: To examine the frequency, nature and determinants of pharmacy compounded medicines in Dutch community pharmacies.

Methods: A prospective nested case-control study comparing prescriptions for pharmacy compounded medicines (cases) with non-pharmacy compounded medicines (controls) was carried out in 79 Dutch community pharmacies. 991 Prescriptions for compounded medicines (cases), dispensed by the pharmacy on a predetermined day in a specific period (29 March until 11 April 2001), and 993 prescriptions for non-compounded medicines (controls) randomly selected on the same day, were studied. The nature and frequency of compounded medicines as well as patient, drug and prescriber related determinants were assessed. In addition, some organisational characteristics, like compounding site and use of protocols, were investigated. Also, the value of compounded medicines in terms of the availability of an industrially compounded equivalent and patient specific reasons, as perceived by the participating pharmacists, was evaluated.

Results: The overall frequency of prescriptions for pharmacy compounded medicines in relation to the total number of prescriptions was 3.4%. This means 12.5 compounded medicines per pharmacy per day on average, but there was a large variation between pharmacies. Excluding the products purchased from specialised compounding companies (28.4%) and the small part of medicines coming from other pharmacies (5.2%), we found an overall frequency of 2.3% of actual compounding in the pharmacy itself. On average, approximately one employee was needed for compounding activities with a large variation between pharmacies. More than 13% of the pharmacists stated that they delivered more than 25% of their compounded medicines to other pharmacies. In 2 pharmacies (2.6%) no actual compounding took place. For 58% of the products manufactured in the pharmacy itself or coming from other pharmacies a (semi-) standardised protocol was used. Compared to non-compounded medicines we found a huge share of dermatological dosage forms among compounded medicines (62.1% versus 5.3%). Oral solutions and ear-nose-throat (ENT) products were also found relatively often. While no ATC class was very pronounced in the control group, the group of dermatologicals was prominently present in the case group (57%) followed by CNS agents (8.4%). The dermatologist was a very strong determinant of compounded medicines compared to GPs (ORadj 12.2 [6.3-23.6]). Patients of 12 years or younger received a significantly higher rate of compounded medicines than persons older than 12 years of age (ORadj 3.4 [2.5-4.8]). Compounding occurred almost twice as often when a medicine was prescribed for the first time compared to a repeat prescription (ORadj 1.8 [1.5-2.2]). In about 63% of the cases the pharmacist judged that an industrially produced medicine could not substitute for the compounded medicine. In about 33% of the compounded products they indicated a patient specific reason. In about 10% this reason concerned a strictly defined pharmaceutical care issue.

Conclusions: Based upon our research, all Dutch community pharmacies compound more than 13,000 medicines per day (2.3% of all prescriptions). They consist mainly of dermatological preparations. Younger children (< 12 yr) receive a significantly higher rate of compounded medicines than other people. At least 1.2 compounded prescriptions per pharmacy per day have a specific pharmaceutical care reason according to the pharmacists.