In 1978, Michael Shepherd gave a Maudsley Bequest Lecture in which it was necessary for him to argue from first principles for the need for epidemiological research within clinical psychiatry (Shepherd, 1978). This paper will take stock of the 20 years since his lecture and point to the new directions for this research approach. Shepherd listed the contributions of psychiatric epidemiology as follows: (a)the completion of the spectrum of disease;(b)the establishment of outcome;(c)the actuarial assessment of morbidity risk;(d)the evaluation of efficacy of treatment; and(e)the conceptual construction of diag nosis/classification.