Viral encephalitis may develop subsequent to viremia, via neuronal spread, or by arthropod vector. Diagnosis often requires invasive studies such as lumbar puncture and brain biopsy. This guideline addresses herpes simplex, rabies, and arbovirus infections of the central nervous system. Clinical trials should be designed according to the availability of approved therapeutic agents. Study designs with an active control (herpesvirus), a placebo control (arbovirus), or no control (rabies virus) are recommended. Outcome should be assessed 4-6 weeks, 4-6 months, and 11-13 months after the completion of therapy. For newborns with encephalitis, outcome should be assessed yearly through the age of 5 years. Assessment of clinical outcome is paramount. © 1992 Oxford University Press.