Lack of correlation between minimal erythema dose and skin phototype in a colombian scholar population

Background: Sun exposure and skin phototype are the most relevant risk factors for skin cancer. Colombia has high levels of ultraviolet radiation during the whole year, therefore, both, high UVI's and outdoor worker's daily activities, in our country are very important risk factors for the development of cutaneous cancer. To date no study has evaluated the usefulness of Fitzpatrick's skin phototype classification in Colombians and its correlation with the minimal erythema dose (MED) and constitutional skin color. Such information is gaining importance in other nations due to the fact that several country's population is becoming more ethnically diverse. Objectives: To determine the skin phototype, accumulated sun exposure, sun protection behavior, MED and phenotype in a Colombian school population. Methods: Last year high school students from the western Antioquia were invited to participate by phone and letter through their respective school directors. A self‐questionnaire was handled to each student. A representative sample of the universe was selected for a medical examination by a dermatologist in order to validate the results of the self‐questionnaire. The constitutional skin color was determined with the chromameter CR 300 Minolta^®. The MED was defined as the minimal dose of UVB being able to induce erythema 24 h later. Results: Eight schools of the area agreed to participate in the study, and a total of 911 students (58% girls and 42% boys) filled‐out the self‐questionnaire. Sun exposure in the majority of individuals was in a level between moderate and very high. Ninety percent of students do not use any sun protection device or cream. Only a 50% of concordance between self‐assessed skin phototype vs. medical skin phototype was found, and the highest concordance corresponded to skin phototype II (82%). There was a marked difference in skin photosensitivity of Colombians compared with reports in Caucasians. We observed a marked overlapping in MED's and L^* values in phototypes II and III. Conclusions: The Fitzpatrick's classification was not useful in Hispanic populations such as ours. Therefore, a new skin‐phototype classification system is required. In our population the constitutional color was a good predictor of the MED but it did not correlate with skin phototype. The self‐assessed questionnaire method was not useful to determine skin cancer risk in our population. The majority of this population has light skin phototypes and is highly exposed to solar UV radiation without proper protection.