Informing the design of a discrete choice experiment for evaluating warfarin pharmacogenetic testing among Mandarin-speaking Chinese warfarin patients in Singapore: a mixed methods analysis.

Dear Editor, Specifi c genetic variants have a substantial effect on warfarin dose response but warfarin pharmacogenetic testing (WPGT) is still not routine clinical practice. Apart from clinical validity, economics, social, ethical and legal implications are also important aspects in the implementation of WPGT. Several studies have revealed high interest in pharmacogenetic testing (PGT) but also concerns over privacy, confi dentiality and cost.1,2 However, these studies were all conducted in non-Asian populations, whom may have different perceptions and attitudes towards PGT from Asians due to cultural differences. On the economic front, the cost-effectiveness of WPGT is inconclusive3 and no cost-benefi t analyses, where willingness-to-pay (WTP) is used to value health benefi ts in monetary terms,4 have been done. The discrete choice experiment (DCE) methodology is an increasingly popular method not just for generating health preferences in healthcare decision-making, but also to elicit WTP.5 In a DCE, individuals are asked to state their preferences between alternative choice sets with each choice set defi ned by a number of attributes, which may include cost and effi cacy in the context of WPGT. To the best of our knowledge, there has been no preference study for WPGT in any population internationally. Although the DCE is relatively intuitive, this methodology is unfamiliar to the Asian population, especially in the fi eld of healthcare. We hereby report the results of a developmental study aimed specifi cally to (i) determine the effectiveness of WPGT educational materials, (ii) identify concerns about WPGT, (iii) identify the most relevant effi cacy attribute(s) for the DCE, and (iv) determine participants’ ability to understand and complete the DCE. Mandarin-speaking Chinese warfarin patients of age ≥ 21 years were recruited from the anticoagulation clinics at the National University Hospital between April to May 2011 using convenience sampling. Patients with signs of cognitive function problems, as perceived by the interviewer, were excluded. Individual voice-recorded, face-to-face interviews were conducted in Mandarin, using a semistructured interview protocol with the aid of show cards. The interview was divided into several sections: (i) education on WPGT and post-education evaluation, (ii) selection of 1 to 2 (from 5 shortlisted) effi cacy attributes important to WPGT, (iii) a trial DCE (using attributes 3 and 4 (Table 1), and cost), and (iv) post-DCE evaluation. The ability to understand WPGT after education was assessed based on the patients’ ability to explain it in their