Jeroen van de Pol

141 5 Preferences of patients regarding community pharmacy services: A discrete choice experiment Methods Study design A cross-sectional study design using an online questionnaire was used among members of the AMP pharmacy patient panel, with a choice based conjoint (CBC) task on potential future community pharmacies services. Conjoint analyses, such as CBC’s, are increasingly used to elicit patients’ preferences for different types of healthcare interventions and services [12-16]. With the help of CBC’s, researchers are able to quantify for example the (relative) importance of different treatment option or (healthcare) service, by allowing respondents to make a number of trade-offs [17]. In CBC, characteristics of treatment options or services are called attributes (e.g. when going to a community pharmacy, waiting time can be an attribute) that can vary in level (e.g. 5, 15 or 30 minutes). Next to the choice task, the questionnaire contained questions on background characteristics of the participants such as age, gender and the number of medication in use. Development process of the choice based conjoint task The development of the CBC task was done in accordance to guidelines described in the literature [3,18,19]. The first step of CBC development entails defining attributes and levels. To identify developments regarding potential future pharmacy services (attributes in this study), a search was conducted through the grey literature. A total of 15 different developments (see supplementary material) were identified. Subsequently, these were presented to 10 experts with knowledge of the pharmacy profession. The experts scored each development on 1) relevance for the patient, 2) relevance for the community pharmacy profession, 3) feasibility to implement in daily community pharmacy practice and 4) the innovative character of the development on a scale of 0-5. Based on the average scores provided by the experts on the 4 domains (with the emphasis on relevance for the patient), a total of 6 developments regarding community pharmacy services were purposely selected to be used as attributes in the CBC. Three of themwere primarily aimed at convenience and three that could be described as a potential CPS (see supplementary material). Adding a cost attribute was not deemed relevant, as in the Netherlands, costs of healthcare services are usually reimbursed by healthcare insurance companies. Therefore, adding a cost attribute could be confusing and become the dominant attribute [20].