%0 Journal Article %A Richard Martin %A Alex Dima %A Annie Burden %A Emilio Pizzichini %A Guy Brusselle %A Janet Holbrook %A Nikolaos G. Papadopoulos %A Todor Popov %A Gene Colice %A Helen Reddel %A Julie von Ziegenweidt %A Alison Chisholm %A David Price %T REG endpoint validation: Association between database and patient-reported exacerbations %D 2014 %J European Respiratory Journal %P P3806 %V 44 %N Suppl 58 %X Aim: Patient-reported exacerbations (PRE) can be hard to capture. We assessed whether a database exacerbation measure (DBE) reliably reflected the patient experience.Methods: We used a UK research database to select clinical and asthma questionnaire records of patients aged 18–60yrs with an asthma diagnosis, no other chronic respiratory diseases and not receiving maintenance oral steroids. DBEs were identified in the clinical records as: a course of oral steroids with evidence of respiratory review (OS) and/or asthma-related hospital admission or emergency room (ER) attendance. PREs over the same period were identified as a positive questionnaire response to any of the 3 questions: In the last 12 months, have you (i)Received courses of OS for worsening asthma? (ii)Been admitted to hospital due to asthma? (iii)Attended ER for asthma? The strength of association between the endpoints was assessed using Kendall's Tau-b (τb).Results: Of a total of 1666 patients (62% female, 24% current smokers), 25% reported ≥1 PRE, and 10% had ≥1 DBE. Some co-linearity was found between PRE items (i)&(ii). A moderate association was found between DBE and PRE (τb=0.419; p<0.001), the table shows the relationship between the measures.View this table:Association between PREs & DBEsConclusion: All outcome measures have their limitations. Despite the different sources of measurement error for DBE and PRE definitions (eg. known under-reporting of hospitalisations in UK primary care records), these measures were found to have a moderate positive association. %U