Do doctors contribute to socioeconomic inequalities in health care provision? An audit experiment in Tunisia

In this paper, we explore an important but understudied driver of health inequalities: whether doctors treat patients from different socioeconomic backgrounds differently during a clinical encounter. We design an audit experiment in Tunisia, sending standardised patients with the same symptoms to 130 public and private primary care doctors for consultation. Informed by in-depth qualitative work, we vary the attitude and appearance of the patients so that they appear to be "poor" or "middle-class". We find no evidence that doctors manage patients differently, but they respond to the socioeconomic profile of patients by prescribing fewer expensive drugs and giving out more free drugs to poorer patients. We also show significant differences in communication between patients: doctors are more likely to provide more explanation to richer patients about the diagnosis, the drugs prescribed and the treatment plan. These differences are not explained by time constraints as doctors spent comparable time with both types of patients. To the extent that differences in communication with patients can lead to differences in patients' health decisions, our results suggest that doctors could contribute indirectly to health inequalities.