The Psoriasis Area and Severity Index (PASI) is used by dermatologic investigators to assess psoriasis disease severity. PASI measures are time-consuming and require trained personnel. We developed and evaluated a structured PASI-like instrument for self-measurement of psoriasis disease severity for use in future large-scale epidemiologic studies. The instrument consisted of a silhouette of a body for patients to shade in affected areas and of three modified visual analog scales for recording the redness, thickness, and scaliness of an average lesion. The resulting "patient PASI" score was compared to investigator PASI measurements in 43 patients with mild to severe psoriasis. Patients and investigator also recorded a subjective global severity score. The patient PASI score was highly predictive of the investigator PASI score (r2 = 0.71, p < 0.0001) and the investigator global score (r2 = 0.37, p < 0.0001). The patient global score was also predictive of the investigator PASI (r2 = 0.18, p < 0.0053) and the investigator global score (r2 = 0.10, p < 0.044), but these relationships exhibit wide predicted confidence intervals. Following the initial evaluation by 18.4 +/- 9.1 d of therapy, one-way t tests for paired samples (pre and post) showed significant improvements in the patient PASI (p < 0.003) and investigator PASI (p < 0.05). We conclude that although patients' overall scores may not allow for accurate prediction of clinical disease severity in a single individual, this structured instrument facilitates prediction of psoriasis disease severity in a population.