EUVIMEDEUVIMED entdecken
Vergleichsstudie

Comparing algorithms for estimating glucocorticoid exposure in Danish patients with rheumatoid arthritis: a study using national registers and DANBIO data

Kirsten S Duch, Peremisan Pakyanathan, Tóri Ásbjørn Matras Holm, Martin Bøgsted, Søren Lundbye-Christensen, Lene Wohlfahrt Dreyer, Salome Kristensen

European Journal of Clinical Pharmacology · 2026-12 · 82 (12)

DOI: 10.1007/s00228-026-04214-4 ↗

Kurz gefasst

Vergleichsstudie in der Zeitschrift European Journal of Clinical Pharmacology, online erschienen am 09.11.2026. Verfasst von Kirsten S Duch und 6 weiteren Autor:innen. Beteiligte Einrichtung: Center of Rheumatic Research Aalborg (CERRA), Department of Rheumatology, Aalborg University Hospital, Aalborg, Denmark. k.duch@rn.dk. Hauptthemen laut MEDLINE: Arthritis, Rheumatoid, Glucocorticoids, Algorithms. Die Arbeit nennt 21 Literaturangaben. Gefördert von: Danmarks Frie Forskningsfond, Gigtforeningen.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

PURPOSE: Estimating glucocorticoid exposure in patients with rheumatoid arthritis (RA) is challenging because Danish national registries provide data on redeemed prescriptions but lack information on daily drug dosage and intraarticular injections. This study compares different methods for estimating glucocorticoid exposure in patients with RA, utilizing data from national registries and hospital records. METHODS: Estimated oral glucocorticoid dose was calculated from registry information, using three methods: (reverse) waiting time distribution ((r)WTD), defined daily dose (DDD), and 90 days average dose. Information on physician prescribed glucocorticoid dosage was gathered from hospital records from 585 patients with RA to validate estimated dose, in both an incident, and a mixed incident and prevalent cohort. Validation was compared using percentage relative difference (RD), limit of agreement ratio (LOAR), sensitivity, and specificity. For glucocorticoid injections, binary agreement between hospital records and registries was presented. RESULTS: Among included models, rWTD performed best for incident populations when comparing physician prescribed glucocorticoid exposure, with registry information (RD: -0.07%, LOAR: 14.34, sensitivity: 1.00, specificity: 0.92). In the mixed cohort, DDD (RD: -23.31%, LOAR: 10.99, sensitivity: 1.00, specificity: 0.95) performed well despite 9% of (r)WTD models showed a lower RD. All LOAR were high. Procedure codes in the registries identified 79% of injections in hospital records in the year following first visit, and 70% of injections in the mixed cohort. CONCLUSION: In the incident cohort, rWTD performed well while both WTD and DDD performed well in the mixed cohort. For injections, procedure codes showed high alignment with hospital records.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Arthritis, RheumatoidGlucocorticoidsAlgorithmsHumansDenmarkRegistriesFemaleMaleMiddle AgedAgedAdult

Stichwörter der Autor:innen

Danish national patient registryDrug dosageGlucocorticoidsRheumatoid arthritis

Substanzen

Glucocorticoids

Autor:innen und Institutionen

  • Center of Rheumatic Research Aalborg (CERRA), Department of Rheumatology, Aalborg University Hospital, Aalborg, Denmark. k.duch@rn.dk.
  • Research Data and Biostatistics, Aalborg University Hospital, Aalborg, Denmark. k.duch@rn.dk.
  • Department of Clinical Medicine, Aalborg University, Aalborg, Denmark. k.duch@rn.dk.
  • The DANBIO Registry, Glostrup, Denmark. k.duch@rn.dk.
  • Center of Rheumatic Research Aalborg (CERRA), Department of Rheumatology, Aalborg University Hospital, Aalborg, Denmark.
  • Department of Clinical Medicine, Aalborg University, Aalborg, Denmark.

Förderung

  • Danmarks Frie Forskningsfond
  • Gigtforeningen

Zitieren

APA

Duch, K. S., Pakyanathan, P., Holm, T. Á. M., Bøgsted, M., Lundbye-Christensen, S., Dreyer, L. W., & Kristensen, S. (2026). Comparing algorithms for estimating glucocorticoid exposure in Danish patients with rheumatoid arthritis: a study using national registers and DANBIO data. European Journal of Clinical Pharmacology, 82(12). https://doi.org/10.1007/s00228-026-04214-4

Vancouver

Duch KS, Pakyanathan P, Holm TÁM, Bøgsted M, Lundbye-Christensen S, Dreyer LW, et al. Comparing algorithms for estimating glucocorticoid exposure in Danish patients with rheumatoid arthritis: a study using national registers and DANBIO data. European Journal of Clinical Pharmacology. 2026;82(12). doi:10.1007/s00228-026-04214-4

BibTeX
@article{Duch2026,
  author = {Duch, Kirsten S and Pakyanathan, Peremisan and Holm, Tóri Ásbjørn Matras and Bøgsted, Martin and Lundbye-Christensen, Søren and Dreyer, Lene Wohlfahrt and Kristensen, Salome},
  title = {Comparing algorithms for estimating glucocorticoid exposure in Danish patients with rheumatoid arthritis: a study using national registers and DANBIO data},
  journal = {European Journal of Clinical Pharmacology},
  year = {2026},
  volume = {82},
  number = {12},
  doi = {10.1007/s00228-026-04214-4}
}
Kostenlos starten →