Sir.—We read with interest the article by Saulsbury and Campbell1 entitled "Evaluation of Child Abuse Reporting by Physicians." Saulsbury and Campbell's data provide important information on physician attitudes, which should ultimately help guide the needed development of educational curricula in medical schools, residency training, and continuing education programs. Becuase of the results of a similar survey, we recently conducted in Illinois (unpublished data), several comments seem germane.
It is easy to understand our colleagues' hesitancy to report suspected cases of child abuse and neglect when the diagnoses are uncertain. In other areas of medical problem solving, a physician establishes hypotheses based on empirical suspicions and cues.2 His/her next step is to test the hypotheses using historical information in conjuction with clinical and laboratory data. Closure and therapeutic intervention usually occur only after the physician is reasonably certain of the diagnosis. In cases of child abuse and neglect, however, closure for