Selection of controls for clinical research studies in ophthalmology
B. S. Hawkins
Wilmer Ophthalmological Institute, Johns Hopkins Medical Institutions, Baltimore, MD 21205.
Ophthalmologists are becoming increasingly aware of the need for
well-designed, controlled studies to investigate etiology and treatment of
ocular conditions. While patients with the conditions of interest, or with
characteristics that place them at high risk of eventual development of
such conditions, may be readily identified in the ophthalmology clinic,
identification of appropriate controls to be used for comparison may be
quite difficult. In this article, epidemiologic principles for selection of
controls for retrospective (case-control) and prospective (cohort) studies
are reviewed. Two related studies, a case-control study in which multiple
control groups were employed and a 15-year cohort study of all cases and
controls, provided an opportunity to compare controls selected at random
from an ophthalmology practice with controls selected at random from the
general population. Participation rates were higher among office patients
selected for study, both in the case-control study and in the cohort study.
In the cohort study, office controls were more likely to lose visual acuity
and to develop new degenerative eye conditions. These differences between
control groups selected from two different sources emphasize the need for
careful evaluation of potential groups of controls with respect to biases
that each may bring to interpretation of clinical research findings.
