Are PASI scores as effective for diagnosis and assessment of disease severity as they are in assessing treatment efficacy in clinical trials?