272 Predictors of squamous cell carcinoma and implications for follow-up in high-risk patients in the Veterans Affairs Keratinocyte Carcinoma Chemoprevention (VAKCC) Trial A Misitzis 1,4 , M Beatson 1,4 , S Landow 1,4 , R Lew 3 , H Higgins 1 and M Weinstock 1,2 1 Brown University Department of Dermatology, Providence, Rhode Island, United States, 2 VA Medical Center, Department of Dermatology, Providence, Rhode Island, United States, 3 VA Healthcare System, Boston, Massachusetts, United States and 4 VA Medical Center Department of Dermatoepidemiology, Providence, Rhode Island, United States Invasive squamous cell carcinoma (SCC) of the skin is one of the most common cancers in USA. We sought to determine the predictors for SCCs in a high-risk population of 932 vet- erans in the Veterans Affairs Keratinocyte Carcinoma Chemoprevention (VAKCC) Trial. Par- ticipants enrolled had at least two keratinocyte carcinomas in the 5 years prior to enrollment and were mainly white males with a median age of 71 years. Participants applied either topical fluorouracil cream 5% or vehicle control cream to the face and ears twice daily for up to 4 weeks. The median follow-up duration was 2.8 years. 11.6% developed a new histo- pathology-confirmed SCC during our trial. The factors that independently predicted a new invasive SCC in descending order were: number of actinic keratosis (AKs) at the beginning of the study, number of prior invasive SCCs, number of prior SCC in situ, no college education and body mass index. The risk for a new SCCs in 2 years was 5%, 12% and 20% for patients with 0, 1 and 2 or more prior SCCs, respectively. The risk for a new SCC in 2 years was 3%, 6% and 14% for patients with 0-2, 3-6 and 7 or more AKs at baseline, respectively. Based on our data, if patients are followed biennially (vs. semiannually), the associated cost of visits over 2 years will be $82.49 vs. $329.96, but the potential risk of delayed diagnosis might be substantial if the patient or their family do not notice and evaluate a new lesion promptly. Our results are a first step in individualizing follow-up schedules for cost-effective management. 273 Defining flare in hidradenitis suppurativa: A systematic review J Kirby 2 , B Moore 1 and P Leiphart 1 1 Penn State College of Medicine, Hershey, Pennsylvania, United States and 2 Department of Dermatology, Penn State Hershey, Hershey, Pennsyl- vania, United States An international multi-stakeholder group reached consensus on a core outcome set of do- mains for HS clinical trials and made recommendations for assessment of HS flare. While considerable effort has been put into developing patient-reported and clinician-reported measures for HS, there has been little done to define HS flare. The goal of this study is to systematically review the literature to investigate prior definitions for HS flare. A systematic review on the definitions and context of flares of HS was performed in PUBMED. Papers were included if it was a systematic review, trial, cohort study, case series or case report, or cross- sectional study. Studies were excluded if it was a journalistic review, did not discuss HS, did not used the terms ‘flare’, ‘exacerbation’, ‘relapse’ or ‘recurrence’, or could not be retrieved or translated into English. Overall, 270 papers were identified, Of these, 23 studies mentioned ’flare’ and 4 studies defined ‘flare’, while 13 studies mentioned ’exacerbation’ but none defined it. Two studies defined flare based on symptoms reported by the patient and 3 by clinical exam changes. The studies indicate a flare includes worsening symptoms or increased inflammation from a prior state, either absence of HS or consistent level of HS activity. Control of flares should be one of the outcomes measured in HS clinical trials. Further research will involve patient and clinician input to create a standardized and measurable definition for HS flare. 274 Spray sunscreen - Characterizing real world use and implications for sun protection L Broussard, S Hirner, R Dellavalle, C Dunnick and J Hugh Dermatology, University of Colorado, Denver, Colorado, United States Typical sunscreen application in liquid form has been shown to be less than 2 mg/cm 2 , which is the FDA required density used in testing, and reflective of the labeled sun protection factor (SPF). Lower sunscreen density has been shown to provide less ultraviolet protection. Aerosol sunscreen is growing in popularity, yet no studies have examined trends in self-application of aerosolized sunscreens. The aim of this study was to quantify the density of aerosol spray formulation used in practical application and to measure the percentage lost in the envi- ronment. Fifty eight participants applied aerosol sunscreen to an absorptive pad on their own forearms and completed a questionnaire. The pre- and post- weights of the absorptive me- dium and sunscreen bottle were used to calculate the density of sunscreen application and amount lost to the environment. The mean aerosol sunscreen application density was 1.75 mg/cm 2 , which was higher than previously reported consumer application of densities of lotion. The median amount of sunscreen lost to the environment was 56%. The study was limited by a moderate sample size, application to a limited area (rather than the whole body), and was performed in one geographic location. Self-application of aerosol sunscreen in the average consumer may not meet the FDA recommended density for adequate SPF protection, providing a lower degree of ultraviolet protection. However, application density for aerosol sunscreen is relatively high, suggesting aerosol sunscreen may provide a higher degree of sun protection for the average user when compared to lotion sunscreen. 275 Predictors of basal cell carcinoma and implications for follow up in high-risk patients in the VAKCC Trial M Beatson 1,2 , A Misitzis 1,2 , S Landow 1,2 , H Higgins 1 , R Lew 3 and M Weinstock 1,2 1 Brown University Department of Dermatology, Providence, Rhode Island, United States, 2 VA Medical Center Department of Dermatoepidemiology, Providence, Rhode Island, United States and 3 VA Healthcare System, Boston, Massachusetts, United States Basal cell carcinoma (BCC) is the most common cancer in the United States. We sought to assess risk factors associated with the development of a subsequent BCC by studying a high- risk population of veterans enrolled in the Veterans Affairs Keratinocyte Carcinoma Chemo- prevention Trial (VAKCC) who had at least 2 keratinocyte carcinomas (KCs) in the 5 years prior to enrollment. Furthermore, we sought to evaluate implications of potential follow up schedules. The number of BCCs in the prior 5 years was the most important predictor of subsequent BCC over the next few years. The hazard ratio of those with >5 BCCs in the prior 5 years, compared to those with <2, was 3.1 (95% CI 2.0, 4.7). Age was an additional in- dependent predictor in the multivariate model with a hazard ratio of 1.3 per decade (95% CI 1.1, 1.5). Participants with >3 prior BCCs have a 9%, 24%, 31%, and 40% risk of developing a BCC in 6, 12, 18, and 24 months, respectively. However, participants with 0-1 prior BCCs have a 3%, 7%, 15%, and 19% risk of developing a BCC during the same intervals. We considered these differences in risk of future BCCs and implications of delayed diagnosis, including morbidity of a larger lesion and associated costs of treatment to evaluate follow up. For example, if participants with >3 prior BCCs, who have a 40% chance of developing a BCC in 2 years, visit once every 6 months for 2 years or once in 2 years, the associated costs are $329.96 and $82.49, respectively. This difference in cost must be balanced against the slow growth rate of BCCs and the likelihood that a new BCC would be noted by a patient before a scheduled follow up, prompting an earlier visit. Stratifying the risk of these patients and evaluating different schedules is an important first step toward creating evidence-based guidelines for optimal management of patients with prior BCCs. 276 U.S. regional differences in psoriasis care: Healthcare utilization, costs, and access to biologics K Nguyen 1 , KK Wu 2 , C Read 3 and AW Armstrong 3 1 University of Central Florida, Orlando, Florida, United States, 2 Frank H. Netter MD School of Medicine at Quinnipiac University, North Haven, Connecticut, United States and 3 USC Keck School of Medicine, Los Angeles, California, United States Regional differences in the U.S. in healthcare utilization and costs among psoriasis patients are unknown. However, patient location may critically impact their access to care. The objective of this study is to compare healthcare utilization, costs, and biologic medication use among patients with psoriasis in the four U.S. census regions: Northeast, Midwest, South, and West. We performed a cross-sectional study using the Medical Expenditure Panel Survey from the years 1996 to 2015. Among the 17,014,231 (weighted) psoriasis patients in the U.S. pooled during the 20-year period, 3,351,104 (19.7%) resided in the Northeast, 4,110,472 (24.2%) resided in the Midwest, 6,010,346 (35.3%) resided in the South, and 3,542,309 (20.8%) resided in the West. The average number of ambulatory visits and ambulatory costs per person per year were 3.8 visits and $1,320 in the Northeast, 2.9 visits and $387 in the Midwest, 3.0 visits and $725 in the South, 5.0 visits and $943 in the West, and 3.5 visits and $803 in the U.S. overall. Compared to the remainder of the county, more psoriasis patients in the South reported using biologic medications (9.87% vs 6.78%, p¼0.045) and had 60% fewer emergency room (ER) visits (p¼0.039), after adjustment for age, gender, race, ethnicity, and insurance status. Compared to the rest of the country, Midwest psoriasis patients had 23% fewer ambulatory visits and incurred less ambulatory costs ($387 vs $934/person/year, p<0.001). Exploratory analysis showed that those with no ambulatory visits in the Midwest had significantly more ER visits than those with one or more ambulatory visits (0.0293 vs 0.0007 visits/person/year, p<0.001). In conclusion, regional differences exist in healthcare utilization, costs, and biologic use. Patients in the South used more biologics and had fewer ER visits than the rest of the country. Midwest psoriasis patients had fewer ambulatory visits and costs compared to the rest of the country. 277 Heritability of tanning addiction: A prospective twin concordance study R Tripathi, KD Knusel, HH Ezaldein, JS Bordeaux and JF Scott Dermatology, Case Western Reserve University and University Hospitals Cleveland Medical Center, Cleveland, Ohio, United States Background: Tanning addiction is a dermatologic-psychiatric disorder with a defined ultra- violet radiation-dependent mechanism and carcinogenic sequelae. Although twin studies show contributions of both environmental and genetic factors for various addictions, no research exists regarding the heritability of tanning addiction. Our objective was to estimate concordance rates and heritability of tanning addiction in a twin cohort. Methods: The validated tanning-modified Diagnostic and Statistical Manual of Mental Disorders, 4 th edition (DSM-IV-TR) scale was used to evaluate tanning addiction in twins at the largest annual twin gathering in the world (Twinsburg, Ohio, 2018). We calculated probandwise concordance rates and tetrachoric correlations with 95% confidence intervals (CIs) to measure the degree of concordance for tanning addiction. Structural equation and liability threshold models were applied to estimate the liability of additive genetic effects (heritability) and unique environ- mental effects of tanning addiction. Results: In total, 147 dizygotic (DZ) and 24 monozygotic (MZ) twin pairs were included (n¼342). 33.0% of the study population met criteria for tan- ning addiction. 34 DZ and 4 MZ pairs were concordant for tanning addiction. Probandwise concordance for MZ pairs was 0.70 (95% CI:0.60-0.80) and for DZ pairs was 0.50 (95% CI 0.20-0.80). Tetrachoric correlation for MZ pairs was 0.75 (95% CI:0.56-0.86) and for DZ pairs was 0.38 (95% CI:0.30-0.45). Controlling for age and gender, 75.4% of tanning addiction was inherited (95% CI:60.6-90.2%) and 24.6% was due to unique environmental effects (95% CI: 9.8-39.4%). Conclusion: The estimated heritability of tanning addiction (75.4%) indicates a substantial genetic component congruent with that of other addictive disorders including alcoholism (70.6%) and nicotine dependence (65.0%). Defining the heritability of tanning addiction is important to better understand its multifaceted pathogenesis and develop tar- geted interventions to reduce its morbidity and mortality. Epidemiology | ABSTRACTS www.jidonline.org S47