A study examining inter-rater and intrarater reliability of a novel instrument for assessment of psoriasis: the Copenhagen Psoriasis Severity Index

Background There is a perceived need for a better method for clinical assessment of the severity of psoriasis vulgaris. The most frequently used system is the Psoriasis Area and Severity Index (PASI), which has significant disadvantages, including the requirement for assessment of the percentage of skin affected, an inability to separate milder cases, and a lack of linearity. The Copenhagen Psoriasis Severity Index (CoPSI) is a novel approach which comprises assessment of three signs: erythema, plaque thickness and scaling, each on a four-point scale (0, none; 1, mild; 2, moderate; 3, severe), at each of 10 sites: face, scalp, upper limbs (excluding hands and wrists), hands and wrists, chest and abdomen, back, buttocks and sacral area, genitalia, lower limbs (excluding feet and ankles), feet and ankles. Objectives To evaluate the inter-rater and intrarater reliability of the CoPSI and to provide comparative data from the PASI and a Physician's Global Assessment (PGA) used in recent clinical trials on psoriasis vulgaris. Methods On the day before the study, 14 dermatologists (raters) with an interest in psoriasis participated in a detailed training session and discussion (2.5 h) on use of the scales. On the study day, each rater evaluated 16 adults with chronic plaque psoriasis in the morning and again in the afternoon. Raters were randomly assigned to assess subjects using the scales in a specific sequence, either PGA, CoPSI, PASI or PGA, PASI, CoPSI. Each rater used one sequence in the morning and the other in the afternoon. The primary endpoint was the inter-rater and intrarater reliability as determined by intraclass correlation coefficients (ICCs). Results All three scales demonstrated 'substantial' (a priori defined as ICC > 80%) intrarater reliability. The inter-rater reliability for each of the CoPSI and PASI was also 'substantial' and for the PGA was 'moderate' (ICC 61%). The CoPSI was better at distinguishing between milder cases. Conclusions The CoPSI and the PASI both provided reproducible psoriasis severity assessments. In terms of both intrarater and inter-rater reliability values, the CoPSI and the PASI are superior to the PGA. The CoPSI may overcome several of the problems associated with the PASI. In particular, the CoPSI avoids the need to estimate a percentage of skin involved, is able to separate milder cases where the PASI lacks sensitivity, and is also more linear and simpler. The CoPSI also incorporates more meaningful weighting of different anatomical areas.