• Vera Vennedey, Sonja Hm Derman, Mickaël Hiligsmann, Daniele Civello, Anja Schwalm, Astrid Seidl, Fülöp Scheibler, Stephanie Stock, Michael J Noack, and Marion Danner.
• Institute for Health Economics and Clinical Epidemiology, University Hospital Cologne (AöR), Cologne, Germany, vera.vennedey@uk-koeln.de.
• Patient Prefer Adher. 2018 Jan 1; 12: 2437-2447.

Background And PurposeThe German Institute for Quality and Efficiency in Health Care (IQWiG) previously tested two preference elicitation methods in pilot projects and regarded them as generally feasible for prioritizing outcome-specific results of benefit assessment. The present study aimed to investigate the feasibility of completing a discrete choice experiment (DCE) within 3 months and to determine the relative importance of attributes of periodontal disease and its treatment.Patients And MethodsThis preference elicitation was conducted alongside the IQWiG benefit assessment of systematic treatments of periodontal diseases. Attributes were defined based on the benefit assessment, literature review, and patients' and periodontologists' interviews. The DCE survey was completed by patients with a history of periodontal disease. Preferences were elicited for the attributes "tooth loss within next 10 years", "own costs for treatment, follow-up visits, re-treatment", "complaints and symptoms", and "frequency of follow-up visits". Patients completed a self-administered questionnaire including 12 choice tasks. Data were analyzed using a random parameters logit model. The relative attribute importance was calculated based on level ranges.ResultsWithin 3 months, survey development, data collection among 267 patients, data analysis, and provision of a study report could be completed. The analysis showed that tooth loss (score 0.73) was the most important attribute in patients' decisions, followed by complaints and symptoms (0.22), frequency of follow-up visits (0.02), and costs (0.03) (relative importance scores summing up to 1).ConclusionA preference analysis performing a DCE can be generally feasible within 3 months; however, a good research infrastructure and access to patients is required. Outcomes used in benefit assessments might need to be adapted to be used in preference analyses.

6 keywords...

hide…