INTRODUCTION: While the care of HIV-positive patients, including the detection and management of comorbidities, has historically been provided in HIV specialist outpatient clinics, recent years have seen a greater involvement of non-HIV specialists and general practitioners (GPs). The aim of this study is to determine whether patients would prefer to see their GP or HIV physician given general symptoms, and to understand what aspects of care influence their preferences. METHODS/ANALYSIS: We have developed and piloted a discrete choice experiment (DCE) to better understand patients' preferences for care of non-HIV-related acute symptoms. The design of the DCE was informed by our exploratory research, including the findings of a systematic literature review and a qualitative study. Additional questionnaire items have been included to measure demographics, service use and experience of non-HIV illnesses and quality of life (EQ5D). We plan to recruit 1000 patients from 14 HIV clinics across South East England. Data will be analysed using random-effects logistic regression and latent class analysis. ORs and 95% CIs will be used to estimate the relative importance of each of the attribute levels. Latent class analysis will identify whether particular groups of people value the service attribute levels differently. ETHICS/DISSEMINATION: Ethical approval for this study was obtained from the Newcastle and North Tyneside Research Ethics Committee (reference number 14/NE/1193). The results will be disseminated at national and international conferences and peer-reviewed publications. A study report, written in plain English, will be made available to all participants. The Patient Advisory Group will develop a strategy for wider dissemination of the findings to patients and the public.