To validate the International Classification of Diseases, 10th revision (ICD-10) diagnostic code L12.0 for bullous pemphigoid (BP) we have performed a retrospective register study based on electronic health records (EHRs) of Oulu University Hospital (OUH) and Helsinki University Hospital (HUH) in Finland. All available data on BP diagnostics were collected from every patient with at least one L12.0 diagnostic code in their EHRs between January 1st, 2009, and December 31st, 2019. BP was confirmed or ruled out based on clinical, immunopathological and serological findings. L12.0 diagnosis was considered to be present if clinical features were characteristic of BP and direct IF was positive; if direct IF was negative, positive BP180 ELISA or positive indirect immunofluorescence was required to the diagnosis. At least one L12.0 was found in the EHRs of 1225 patients and 900 of them had correct diagnosis of BP, thus positive predictive value (PPV) for BP was 73.5% (95% CI 71.0-75.9). For the OUH subpopulation we collected the data of specialty for the registration of L12.0 and performed a receiver operating characteristics (ROC) analysis. The area under curve (AUC) in the ROC analysis was 0.800 (95% CI 0.729-0.870) for L12.0 registered in any specialty and 0.943 (95% CI 0.907-0.979) in the dermatology unit. At least two diagnoses registered in dermatology unit proved as the optimal cut-off value to recognize BP patients in a dataset like ours (PPV: 96.3% (95% CI 94.0-98.6), sensitivity: 92.0% (95% CI 88.7-95.3), specificity: 88.5% (95% CI 84.7-92.4)). We conclude that L12.0 registered in the dermatology unit is significantly more accurate to recognize patients with true BP in a large-scale epidemiological data than L12.0 registered in any specialty. The inclusion of patients with at least two L12.0 codes registered in dermatology most likely selects the patients with true BP from an EHR based data.

To validate the International Classification of Diseases, 10th revision (ICD-10) diagnostic code L12.0 for bullous pemphigoid (BP) we have performed a retrospective register study based on electronic health records (EHRs) of Oulu University Hospital (OUH) and Helsinki University Hospital (HUH) in Finland. All available data on BP diagnostics were collected from every patient with at least one L12.0 diagnostic code in their EHRs between January 1st, 2009, and December 31st, 2019. BP was confirmed or ruled out based on clinical, immunopathological and serological findings. L12.0 diagnosis was considered to be present if clinical features were characteristic of BP and direct IF was positive; if direct IF was negative, positive BP180 ELISA or positive indirect immunofluorescence was required to the diagnosis. At least one L12.0 was found in the EHRs of 1225 patients and 901 of them had correct diagnosis of BP, thus positive predictive value (PPV) for BP was 73.6% (95% CI 71.0-76.0). For the OUH subpopulation we collected the data of specialty for the registration of L12.0 and performed a receiver operating characteristics (ROC) analysis. The area under curve (AUC) in the ROC analysis was 0.800 (95% CI 0.729-0.870) for L12.0 registered in any specialty and 0.943 (95% CI 0.907-0.979) in the dermatology unit. At least two diagnoses registered in dermatology unit proved as the optimal cut-off value to recognize BP patients in a dataset like ours (PPV: 96.3% (95% CI 94.0-98.6), sensitivity: 92.0% (95% CI 88.7-95.3), specificity: 88.5% (95% CI 84.7-92.4)). We conclude that L12.0 registered in the dermatology unit is significantly more accurate to recognize patients with true BP in a large-scale epidemiological data than L12.0 registered in any specialty. The inclusion of patients with at least two L12.0 codes registered in dermatology most likely selects the patients with true BP from an EHR based data.

ESDR137 - Visit to Dermatology Unit Increases Positive Predictive Value of Bullous Pemphigoid Diagnosis

The scientific meeting of the European Society for Dermatological Research (ESDR) is held yearly, usually in September. With an international attendance of more than 1000 skin scientists from around the globe, the four-day event includes short oral presentations from peer-reviewed abstracts, keynote lectures from international experts, sponsored symposia, satellite programs, educational events as well as an important focus on posters, networking and building collaborations amongst scientists. For more information about this year’s meeting visit our website.