Do general practitioners know patients' preferences? An empirical study on the agency relationship at an aggregate level using a discrete choice experiment.

OBJECTIVES This study investigated whether general practitioners (GPs) know patients' preferences regarding a number of organizational characteristics in general practice (i.e., waiting time on the telephone, opening hours, waiting time to the appointment, distance to the general practice, waiting time in the waiting room, consultation time, and whether the GP or assisting personnel performs routine tasks) to examine whether there is a basis for improving the agency relationship at an aggregate level. DATA A total of 698 respondents from the Danish population and 969 GPs answered the questionnaire in May and September 2010. METHODS In a discrete choice experiment, GPs and patients made both forced and unforced choices, allowing us to explore the congruence of preferences 1) when patients must choose a new GP and 2) when they can stay with their current GP. RESULTS Results show that in the forced choice, preferences are seen to differ. In the unforced choice also, preferences differ--mainly because GPs overestimate their own importance to the patients. Rank orders, however, are similar for both GPs and patients. CONCLUSIONS It is concluded that GPs do not have a precise knowledge of patients' preferences. However, in the unforced choice, GPs do know on which attributes to compete although they underestimate the necessity of competition. The overall conclusion is that there is room for improving the agency relationship in the organization of general practice.