371-P: Different Definition Wording of Severe Hypoglycemia (SH) Yields Variable Prevalence and Incidence Rates

Introduction (2) a large diabetes community website (diabetes.co.uk); (3) a hypoglycemia awareness assessing instrument (HypoA-Q); and (4) the IHSG 2017 consensus. Analyses included McNemar test, kappa statistic, pair-wise Wilcoxon signed-rank test and Spearman correlation coefficient calculation. Results: 1,580 participants (52% women, medianIQR age 4333—58y, diabetes duration 2412—37y, 87% non-Hispanic Caucasian) completed the survey. SH prevalence with the 2013 IHSG definition wording was higher than the website, HypoA-Q and 2017 IHSG definition wording (Figure), with pairwise kappas ranging from fair to good (website definition .75; HypoA-Q .72; the 2017 IHSG definition .69). Compared to the 2013 IHSG definition wording, the website, HypoA-Q and 2017 IHSG definition wording yielded lower incidence. The 2013 IHSG definition wording correlated strongly with the website (rs=.73), HypoA-Q (rs=.72) and moderately with the 2017 IHSG definition wording (rs=.61) (P.001 for all). Conclusion: Reported SH prevalence and incidence vary based on definition wording. Disclosure Y. Lin: None. W. Ye: None. E. Hepworth: None. L. Ang: None. N. de Zoysa: None. S.A. Amiel: Advisory Panel; Vertex Pharmaceuticals Incorporated. Other Relationship; Sanofi. S.J. Fisher: None. Funding National Institutes of Health (U01DK135111, K23DK129724)