Users' guides to the medical literature - XXIV. How to use an article on the clinical manifestations of disease

Clinicians rely on knowledge about the clinical manifestations of disease t o make clinical diagnoses. Before using research on the frequency of clinic al features found in patients with a disease, clinicians should appraise th e evidence for its validity, results, and applicability. For validity, 4 is sues are important-how the diagnoses were verified, how the study sample re lates to all patients with the disease, how the clinical findings were soug ht, and how the clinical findings were characterized, Ideally, investigator s will verify the presence of disease in study patients using credible crit eria that are independent of the clinical manifestations under study. Also, ideally the study patients will represent the full spectrum of the disease , undergo a thorough and consistent search for clinical findings, and these findings will be well characterized in nature and timing, The main results of these studies are expressed as the number and percentag es of patients with each manifestation. Confidence intervals can describe t he precision of these frequencies. Most clinical findings occur with only i ntermediate frequency, and since these frequencies are equivalent to diagno stic sensitivities, this means that the absence of a single finding is rare ly powerful enough to exclude the disease, Before acting on the evidence, c linicians should consider whether it applies to their own patients and whet her it has been superseded by new developments, Detailed knowledge of the c linical manifestations of disease should increase clinicians' ability to ra ise diagnostic hypotheses, select differential diagnoses, and verify final diagnoses.