Actinic keratoses (AKs) are frequently treated in the U.S., impacting an estimated 40 million people in 2004 and costing over $1 billion annually. AKs are a major public health concern because of their high prevalence, substantial cost, and potential for malignant transformation to keratinocyte carcinoma (KC). In this analysis, predictors of AK count were explored using pre-randomization baseline data from two large randomized trials of veterans with multiple prior KCs (n=932 and n=1131). Multivariate analyses were conducted to elucidate associations between AK count and several demographic and health related factors. In both trials, greater baseline AK count on the face/ears was strongly associated (p≤0.01) with older age, lower latitude, male sex, greater sun sensitivity, previous 5-FU use, and a higher number of prior invasive SCCs, but was not associated with number of prior BCCs. Additionally, mean baseline AK count was higher on the left side of the face/ear compared to the right, which may be explained by increased UV radiation to the left side while driving. Risk factors for actinic keratoses in a high-risk population are particularly important as actinic keratoses in this group are more likely to progress to keratinocyte carcinoma. Recognizing predictors of AK count in individuals at high risk for KC may help providers tailor AK prevention and treatment efforts. This may in turn lower the risk of KC—perhaps a more important goal—as the keratinocytic dysplasia that gives rise to malignancy, and sometimes appears as an AK, is what actually threatens patient health.

Keratinocyte carcinoma (KC) is the most common malignancy in the U.S. Because KC affects many people but has a low mortality rate, it is important to consider the impact of KC on skin-related quality of life (QoL). We aimed to determine if KCs, actinic keratoses (AKs), and several demographic and health-related factors were associated with skin- related QoL in those at high risk for KC. We used data from a double-blind, placebo- controlled trial in which 932 veterans with multiple prior KCs were randomized to apply topical 5-fluorouracil (5-FU) or vehicle control cream to the face/ears. Skin-related QoL was measured at baseline, 12, 24, and 36 months using Skindex-29 and Skin Cancer Index and participants received skin examinations semiannually. At baseline, worse skin- related QoL was strongly associated (p≤0.01) with younger age and greater functional impairments (e.g. inability to bathe, walk). It was also associated with higher comorbidities, increased sun-sensitivity and prior 5-FU use, but almost entirely unrelated to baseline AKs and prior KCs. However, participants who developed new KCs or more AKs during the trial had worse skin-related QoL, particularly in year 1. Our ability to detect these relationships in longitudinal analyses but not in cross-sectional analyses may indicate the importance of the more precise control of potential confounding factors that is inherent in comparing each individual to their own prior state. These findings suggest that additional KCs and AKs may worsen skin-related QoL in a high-risk population and individuals with worse overall health may need more dermatologic attention, not less.