124 The risk of liver disease in psoriasis patients: A population-based cohort SK Grewal 1 , A Ogdie 2 , J Takeshita 1,4 , DB Shin 1 , RM Carr 3 and JM Gelfand 1,4 1 Dermatology, University of Pennsylvania, Philadelphia, PA, 2 Medicine/Division of Rheumatology, University of Pennsylvania, Philadelphia, PA, 3 Medicine/Division of Gastroenterology, University of Pennsylvania, Philadelphia, PA and 4 Center for Clinical Epidemiology and Biostatistics, University of Pennsylvania, Philadelphia, PA Previous studies have suggested that psoriasis is associated with liver disease, possibly related to metabolic co-morbidities, alcohol consumption, treatment toxicity, and chronic inflam- mation. The objective of this study was to determine if psoriasis increases risk for incident liver disease after controlling for traditional risk factors for liver disease. We conducted a population-based cohort study in the United Kingdom using The Health Improvement Network (THIN) data from 1994-2014 to investigate the risk of liver disease in psoriasis patients. Patients with a history of liver disease were excluded from the analyses. Diagnostic codes were used to identify psoriasis patients. Patients who received systemic or phototherapy were classified as having moderate to severe psoriasis. Each exposed patient had up to 5 matched controls. Cox proportional hazards models were used to estimate hazard ratios with 95% confidence intervals. Psoriasis patients were at a significantly higher risk for developing liver disease (mild PsO: aHR 1.38 CI 1.30-1.47; severe PsO: aHR 1.74 CI 1.43-2.13) and cirrhosis (mild PsO: aHR 1.64 CI 1.42-1.90; severe PsO: aHR 2.08 CI 1.30-3.31) after adjusting for known risk factors including age, sex, BMI, drinking habits, smoking habits, NSAID use, and diabetes. Psoriasis patients also had an increased risk for fatty liver disease in age/sex adjusted models (mild PsO: HR 1.17 CI 1.07-1.27; severe PsO: HR 3.06, CI 2.44- 3.85). Severe psoriasis conferred the highest risk for all 3 outcomes. These findings indicate that psoriasis patients have an increased risk of development of incident liver disease. This risk is particularly increased in patients with severe psoriasis and is independent of postulated risk factors for liver disease including obesity. 125 Disagreement between physician- and patient-reported disease severity in adults with a history of moderate-to-severe atopic dermatitis W Wei 1 , P Anderson 2 , A Gadkari 3 , S Blackburn 2 , R Moon 2 , J Piercy 2 , S Shinde 1 , J Gomez 3 and E Ghorayeb 1 1 Sanofi, Bridgewater, NJ, 2 Adelphi Real World, Bollington, United Kingdom and 3 Regeneron, Tarrytown, NY Since there is limited understanding on how atopic dermatitis (AD) severity is rated in routine clinical practice, this study evaluated agreement between physician- and patient-rated AD severity and factors associated with disagreement. Data were from the 2014 Adelphi US AD Disease Specific Program, a cross-sectional study of physicians (primary care, dermatologists, and allergists) and their patients with a history of moderate-to-severe AD; patients voluntarily completed a questionnaire. Current disease severity (mild, moderate, severe) was subjectively rated by their treating physician and independently by the patients themselves. Bivariate analyses characterized rating agreement and multinomial logistic regression identified factors independently associated with disagreement. Included patients (N¼678) were 54.4% female, 77.4% white, mean age 39.3 yrs. Only moderate agreement was observed (kappa¼.52): compared with physician ratings, patient-rated severity was higher in 11.2% of cases, lower in 20.2%, matched in 68.6%. Patterns of agreement between physician and patient ratings showed no differences based on physician specialty (P¼ .68), objective severity rating assessed by Eczema Area and Severity Index (P¼.35), % body surface area affected (P¼.99), or current treatment with systemic immunosuppressants (P¼.92). Patients were more likely to rate AD as more severe than their physicians if they had worse quality of life, assessed using the Dermatology Life Quality Index (relative risk ratio [RRR] 1.04, 95% CI 1.00e1.08; P¼.05). Physicians were more likely to rate higher AD severity if the patient reported more sleep disturbance (RRR¼1.71, 95% CI 1.01e2.89; P¼ .04). These results showed that almost one- third of patients rated their AD severity differently from their treating physicians. This was primarily driven by patient-reported outcomes, suggesting the importance of including the patient perspective in assessment of AD severity. 126 Clinical implications of negative direct immunofluorescence (DIF) results in Mucous Membrane Pemphigoid (MMP) with ocular disease M Labowsky 2 , C Shieh 1 , S Stinnett 1 , M Daluvoy 1 and R Hall 2 1 Ophthalmology, Duke University, Durham, NC and 2 Dermatology, Duke University, Durham, NC The clinical diagnosis of ocular MMP is often confirmed by biopsy demonstrating linear IgG, IgA or C3 at the dermal-epidermal junction on DIF. However, some patients with clinically typical disease have fibrin alone or no immune-reactants on DIF. In a retrospective cohort study of patients with MMP with ocular involvement, we investigated whether there was a difference between patients with positive biopsies and those with biopsy-negative but clini- cally typical disease. We examined 53 patients with a clinical diagnosis of ocular MMP, DIF results, and >1 year of follow-up evaluated between 1996 and 2015. Subjects with pseu- dopemphigoid or history of SJS/TEN were excluded. We documented visual acuity, medical history (including malignancy and autoimmune disease) and physical exam findings. Foster’s Stage and MMP Disease Activity Index (MMPDAI) were assessed at initial presentation and at 18 months in 49 subjects. 14 of 25 (56%) subjects with ocular disease alone had positive ocular biopsy results while 26 of 28 (92.3%) with combined ocular, skin or other mucosal disease had positive biopsies (10 mucosal, 7 ocular, 6 skin, and 3 multiple biopsy sites). The proportion of subjects with a positive biopsy was significantly lower in those with ocular disease alone than in those with other tissues involved (p¼0.004). No significant difference was seen in the severity of initial presentation or frequency of autoimmunity/malignancy between subjects with positive and negative biopsies or between subjects with ocular disease alone and those with involvement of mucosa or skin. In addition, no significant difference was seen in Foster’s Stage, MMPDAI, and vision loss at presentation and at 18 months be- tween these groups. These findings demonstrate that clinically diagnosed MMP patients with isolated ocular disease frequently have negative DIF findings yet similar disease presentation and progression and suggest that they should be managed similarly to those subjects with immunologically proven disease. 127 Diagnostic features and treatment responses in non-bullous and pre-bullous pemphigoid V Laniosz 3 , T Hocker 1 , JA Cappel 2 , JS Lehman 3 and MJ Camilleri 3 1 Advanced Dermatologic Surgery, Overland Park, KS, 2 Surgical Dermatology Group, Birmingham, AL and 3 Dermatology, Mayo Clinic, Rochester, MN Non-bullous and pre-bullous pemphigoid are increasingly becoming recognized disease entities typically in elderly patients with recalcitrant dermatitis or pruritus. We sought to determine our experience at a tertiary referral center with atypical pemphigoid by performing a retrospective review of all cases of pemphigoid seen from 2006-2010. After fulfilling obligatory criteria of positive direct immunofluorescence with IgG and C3 in conjunction with serologic evidence of circulating autoantibodies basement membrane zone antigens, we identified 26 cases of pemphigoid presenting with bullae, 28 cases with a pre-bullous pro- drome, and 7 cases of non-bullous pemphigoid. Data were collected on patient de- mographics, time to diagnosis, presenting symptoms, biopsy and laboratory findings, as well as treatment responsiveness. Not surprisingly, diagnosis of both pre-bullous and non-bullous pemphigoid was significantly delayed compared to cases presenting with bullae. Differences in presenting symptoms, indirect immunofluorescence titers, BP180 and 230 positivity and level, eosinophil count, and treatment response were observed amongst the three groups. 128 Cumulative oral corticosteroid use increases risk of glucocorticoid-related adverse events in patients with pemphigus D Wormser 2 , DM Chen 3 , PG Brunetta 3 , MS Broder 1 , E Chang 1 and SR Reddy 1 1 Partnership for Health Analytic Research, LLC, Beverly Hills, CA, 2 F. Hoffmann-La Roche Ltd, Basel, Switzerland and 3 Genentech, Inc., South San Francisco, CA The purpose of this study was to investigate the risk of glucocorticoid-related adverse events (GAE) among pemphigus patients taking oral corticosteroids (OCS). We analyzed 2010-2014 commercial claims data to examine risk of GAE and associated costs among patients with a newly diagnosed episode of pemphigus (ICD-9-CM: 694.4), defined as no pemphigus diag- nosis within 6 months before the first date of diagnosis. We used Cox regression models to estimate risk of GAE associated with cumulative OCS exposure (daily aggregated prednisone- equivalent dose) during 1 year follow-up (median¼785 days observation), adjusting for age, gender, and comorbidity. Annualized costs were compared between patients with and without GAE using t-tests. We identified 644 patients with a new pemphigus episode (mean age¼59.5 years, 56.1% female). Mean number of chronic conditions at baseline was 2.7. Mean total and daily OCS dose in the 1st follow-up year was 3.9g and 29.0mg, respectively. Rate of GAE was 0.46 events per patient year. After adjusting, the hazard ratio (HR) per 1g OCS exposure was 1.01 (P¼0.03), i.e. each additional 1g of OCS was associated with a 1% higher risk of GAE. Risk of cataract (HR: 1.02; P<0.001) and fracture (HR: 1.01; P¼0.03) increased with OCS exposure. An observed increase in risk of infection, osteoporosis, ne- crosis of bone, psychosis, glaucoma, and type 2 diabetes was not statistically significant. Mean annualized healthcare costs for patients with GAE exceeded those of patients without ($38,445 vs. $23,113; P<0.01). Our study demonstrates that pemphigus patients exposed to high-dose OCS have a higher risk of cataract and fracture and that patients with GAE incur greater healthcare costs. Clinical management focused on steroid-free remissions or lower OCS exposure may reduce the detrimental effects demonstrated in this study. 129 Extent and consequences of inadequate disease control among adults with a history of moderate-to-severe atopic dermatitis W Wei 1 , P Anderson 2 , A Gadkari 3 , S Blackburn 2 , R Moon 2 , J Piercy 2 , S Shinde 1 , J Gomez 3 and E Ghorayeb 1 1 Sanofi, Bridgewater, NJ, 2 Adelphi Real World, Bollington, United Kingdom and 3 Regeneron, Tarrytown, NY Atopic Dermatitis (AD) management remains challenging. This study investigated AD control in routine clinical practice and identified factors associated with failure to attain disease control. Data were from the 2014 Adelphi US AD Disease Specific Program, a cross-sectional study of physicians (primary care n¼102; dermatologists n¼75; allergists n¼25) and their patients with history of moderate-to-severe AD (N¼1064; 54% female; 75% white; mean age 40 y), of whom 655 patients voluntarily completed a questionnaire. 58.7% patients (n¼625) had uncontrolled AD as assessed by their physicians, defined as either flaring/deteriorating/ changeable AD or dissatisfaction with current control. With available data, rate of uncon- trolled AD increased with patients’ current AD severity, regardless of whether it was assessed objectively (Eczema Area and Severity Index: moderate, 62.2%; severe, 76.6%) or subjec- tively (physician-rated: moderate, 61.3%; severe, 92.7%; patient-rated: moderate, 64.2%; severe, 84.9%). Uncontrolled AD was seen in 53.4% of patients receiving a systemic immunosuppressant and 83.4% receiving a systemic corticosteroid. Relative to controlled patients, uncontrolled patients reported lower quality of life (Dermatology Life Quality Index mean score 8.1 vs 5.6; P<.001), and greater overall work impairment (22.8% vs 18.4%; P¼.04) among the 407 employed patients who completed the Work Productivity and Activity Impairment questionnaire. Logistic regression showed that independent factors significantly associated with uncontrolled AD were Hispanic ethnicity (odds ratio [OR]¼2.6, 95% CI 1.2e5.7; P¼.015); AD symptoms on head/neck (OR¼1.6, 95% CI 1.2e2.3; P¼.005) or lower limbs (OR¼1.5, 95% CI 1.1e2.1; P¼.013); itch interference with daily living (OR¼2.2, 95% CI 1.5e3.2 P<.0001); sleep disturbance (OR¼2.1; 95% CI 1.3e3.4; P¼.005). This study showed a high rate and substantial impact of uncontrolled AD, suggesting the need for more effective therapies. Clinical Research I: Epidemiology and Patient Outcomes Research | ABSTRACTS www.jidonline.org S23