Response bias occurs when there is a systematic difference between the responses of participants and the true values or behaviors being measured. In epidemiology, this can lead to incorrect conclusions about the prevalence, incidence, or risk factors of diseases.