1. Conferences Overview
    2. 33. Jahrestagung der Gesellschaft für Arzneimittelanwendungsforschung und Arzneimittelepidemiologie (GAA)
    3. Overtreatment among levothyroxine users defined by diagnostic and prognostic thyrotropin (TSH) cut-offs: a population-based study
Logo

33. Jahrestagung der Gesellschaft für Arzneimittelanwendungsforschung und Arzneimittelepidemiologie (GAA)

Gesellschaft für Arzneimittelanwendungsforschung und Arzneimittelepidemiologie (GAA) e. V.
29.-30.10.2026
München

Meeting Abstract

Überbehandlung bei Levothyroxin-Anwenderinnen und -Anwendern anhand diagnostischer und prognostischer Thyrotropin (TSH)-Grenzwerte: Eine bevölkerungsbasierte Studie

Oliver H. F. Scholle - Leibniz Institute for Prevention Research and Epidemiology – BIPS, Bremen, Germany
Till Ittermann - University Medicine Greifswald, Greifswald, Germany
Henry Völzke - University Medicine Greifswald, Greifswald, Germany
Hermann Pohlabeln - Leibniz Institute for Prevention Research and Epidemiology – BIPS, Bremen, Germany
Joachim Feldkamp - Bielefeld University, Bielefeld, Germany
Kathrin Günther - Leibniz Institute for Prevention Research and Epidemiology – BIPS, Bremen, Germany
Nele Friedrich - University Medicine Greifswald, Greifswald, Germany
Matthias Nauck - University Medicine Greifswald, Greifswald, Germany
Lilian Krist - Charité - Universitätsmedizin Berlin, Berlin, Germany
Michael F. Leitzmann - University of Regensburg, Regensburg, Germany
Yong Li - University of Freiburg, Freiburg, Germany
Wolfgang Lieb - Kiel University, Kiel, Germany
Rafael Mikolajczyk - Martin Luther University Halle-Wittenberg, Halle (Saale), Germany
Ilias Moreno Velásquez - Max-Delbrück-Center for Molecular Medicine in the Helmholtz Association (MDC), Berlin, Germany
Nadia Obi - University Medical Center Hamburg-Eppendorf, Hamburg, Germany
Tobias Pischon - Max-Delbrück-Center for Molecular Medicine in the Helmholtz Association (MDC), Berlin, Germany
Börge Schmidt - University Hospital of Essen, Essen, Germany
Ulla T. Schultheiss - University of Freiburg, Freiburg, Germany
Konrad H. Stopsack - Leibniz Institute for Prevention Research and Epidemiology – BIPS, Bremen, Germany
Berit Brandes - Leibniz Institute for Prevention Research and Epidemiology – BIPS, Bremen, Germany

Text

Background/Aim: Many levothyroxine users have thyrotropin (thyroid-stimulating hormone, TSH) levels below the diagnostic reference range (“diagnostic” cut-offs), and are therefore considered overtreated. Evidence from predominantly untreated cohorts has led to a definition of TSH cut-offs based on cardiovascular and all-cause mortality risk (“prognostic” cut-offs). It is unclear how often overtreatment according to prognostic cut-offs occurs. We estimated the prevalence of overtreatment among levothyroxine users based on both cut-offs and identified subgroups with a higher prevalence of overtreatment.

Methods: In the baseline assessment of the German National Cohort (NAKO), a population-based prospective multi-center cohort study, TSH levels were available from 109,997 participants aged 19–74 years. We used study-specific TSH cut-offs to assess the prevalence of overtreatment among levothyroxine users.

Results: Among 13,095 levothyroxine users, 19.3% (95% confidence interval [CI] 18.6–20.0) were overtreated using the diagnostic TSH cut-off (<0.45 mIU/L) and 42.5% (95% CI 41.6–43.3) were overtreated according to the prognostic cut-off (≤0.87 mIU/L). A higher prevalence of levothyroxine overtreatment based on the prognostic TSH cut-off was particularly observed in older age groups (e.g., ≥60 years vs. 20–39 years: adjusted prevalence ratio [aPR] 1.32, 95% CI 1.21–1.44), in users with a long time since first thyroid diagnosis (e.g., ≥10 years vs. ≤1 year: aPR 1.88, 95% CI 1.39–2.54), and in those with higher levothyroxine doses (e.g., >1–1.5 µg/kg/day: aPR 1.58, 95% CI 1.47–1.70; >1.5 µg/kg/day: aPR 2.12, 95% CI 1.96–2.29; each compared with ≤0.67 µg/kg/day).

Discussion/Conclusion: One in five levothyroxine users was overtreated according to the diagnostic TSH cut-off; this prevalence was twice as high when applying the prognostic TSH cut-off. The identification of subgroups with a higher prevalence of overtreatment in this study may help prioritize patients for potential dose reduction or deprescribing of levothyroxine.