A validation study for the classification of the level of severity of malignant skin lesions in cancer patients

A methodological study using a modified Delphi survey was conducted to determine whether or not there was consensus among wound care experts in identifying categories and characteristics essential for classifying malignant skin lesion severity in cancer patients.

Two groups of wound care experts, oncology nurses (ONS) and certified enterostomal therapy nurses (CETN), were sent a demographic data form and two drafts of the malignant skin lesion survey form. Participants were asked to identify descriptors for a specific category domain in the first round of the survey. A subsequent survey form was mailed to the 246 wound care experts in which they were asked to identify the importance of rank for each category domain and appropriate descriptors.

A total of 125 second round surveys were returned (55 ONS and 65 CETN). Items ranked as 5 (extreme importance) and 4 (great importance) on the Likert-type scale by 75% or more of the respondents resulted in inclusion of 28 descriptors on the Malignant Skin Lesion Assessment Instrument.

A t test for independent samples was completed on the second survey results. Agreement existed on the majority of descriptors; two significant differences occurred between groups $(p\le .002),$ and eight type II errors occurred. No significant difference was found for any descriptors on the new instrument. Coefficient alpha for the new instrument is.91.

Results indicated there is consensus among wound care experts relative to descriptor importance for an instrument identifying severity of malignant skin lesions in cancer patients. However, further testing is needed to establish consensus among wound care experts for category titles.