Patient opinion of scarring is multidimensional: An investigation of the POSAS with confirmatory factor analysis
School of Medical and Health Sciences
Scarring is a significant consequence for patients following a burn. Understanding how patients perceive the physiological scar and define scar severity may provide valuable information regarding how the scar influences quality of life after burn. The Patient and Observer Scar Assessment Scale was the first scar assessment tool validated to include the patients’ evaluation of the scars physical qualities, following a burn. Validation studies of this tool have previously been conducted for a discrete scar-site after burn. The aim of this study was to assess the structural validity of the POSAS to capture the patients’ evaluation of the total area of burn scar(s).
Statistical analysis was based on 508 completed POSAS forms from 358 patients. Exploratory factor analysis (EFA) was used initially to identify the number of factors within the tool, then confirmatory factor analysis (CFA) using structural equation modelling explored areas of misfit within each factor and whether the model provided a predicable structure to capture patient perception of scar severity.
The CFA analysis confirmed that a two dimensional model was superior to a unidimensional model when assessing the patient opinion of their total burn scar. The two dimensions were the physical scar (color, stiffness, thickness and irregularity) and the sensory scar (pain and itch). Further strain analysis of the two factor model identified additional domains. Independent factors influenced the perception of color forming a separate subdomain within the physical domain. Color is a visual characteristic, whereas the other three are predominantly tactile characteristics. A significant relationship between thickness and irregularity suggested they may form another subdomain, however further research is required to confirm this. Both pain and itch were recognized as independent, multidimensional latent variables, which require assessment tools with multidimensional structures.
When assessing the entire burn scar, three independent dimensions influence patient perception: (1) the physical scar, (2) pain and (3) itch. Within the physical domain, color formed a visual subdomain separate to a tactile subdomain. Further development of these domains within a high-order multi-dimensional structure is recommended.