Error replication as an instructional strategy in simulation‐based procedural learning
Medical Education · 23.08.2026
DOI: 10.1111/medu.70290 ↗Kurz gefasst
Fachartikel in der Zeitschrift Medical Education, online erschienen am 23.08.2026. Verfasst von Jacob Faurholdt und 3 weiteren Autor:innen. Beteiligte Einrichtung: Copenhagen Academy for Medical Education and Simulation (CAMES), Copenhagen, Denmark. Die Arbeit nennt 32 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Novo Nordisk Fonden, Helsefonden, Gangstedfonden.
Abstract (Originalsprache)
Background
Errors are an inevitable part of clinical learning, yet procedural training commonly emphasises error avoidance. This study examined the effect of deliberate replication of predefined procedural errors (Error Replication; ER), compared with error avoidance training (EAT), on post‐test performance, retention and transfer in surgical simulation.
Methods
In this randomised controlled trial, 70 medical students with no prior simulation‐based osteosynthesis experience completed a pre‐test of simulated dynamic hip screw (DHS) fixation before 60 min of ER or EAT training. ER participants deliberately reproduced predefined, clinically relevant errors, whereas EAT participants used simulator feedback to improve performance and minimise errors across successive iterations. Participants then completed an immediate DHS post‐test, a DHS retention test after 6–10 days and a transfer test involving antegrade femoral nailing. Transfer‐test performance was the primary outcome. Secondary outcomes included performance trajectories across pre‐test, post‐test and retention test; completed DHS iterations; self‐perceived dexterity and spatial awareness; and perceived training efficacy.
Results
Complete data were available for 68 participants. Performance improved substantially over time in both groups ( p < 0.001, η 2 p = 0.606), with no significant group‐by‐time interaction ( p = 0.44). Transfer scores were 52.6 (SD 24.7) for EAT and 51.0 (SD 33.9) for ER ( p = 0.82; d = 0.06, 95% CI − 0.42 to 0.53). No statistically significant differences were detected in the remaining secondary outcomes.
Conclusions
Participants improved following both instructional approaches, although those in the ER group spent most of the training period deliberately reproducing predefined errors. No statistically significant differences between ER and EAT were detected in post‐test performance, retention or transfer. These findings suggest that structured exposure to clinically relevant procedural errors warrants further study as a distinct instructional modality.
Autor:innen und Institutionen
- Copenhagen Academy for Medical Education and Simulation (CAMES), Copenhagen, Denmark.
- Department of Clinical Medicine, Faculty of Health Sciences, University of Copenhagen, Copenhagen, Denmark.
- Department of Orthopedic Surgery, Center for Orthopaedic Research and Innovation (CORI), Næstved, Slagelse and Ringsted Hospitals, Slagelse, Denmark.
- Department of Gynaecology, Fertility, and Obstetrics, Rigshospitalet, Copenhagen, Denmark.
ORCID-Profile: Jacob Faurholdt, Mads Emil Jacobsen, Martin Grønnebæk Tolsgaard, Amandus Gustafsson
Förderung
- Novo Nordisk Fonden
- Helsefonden
- Gangstedfonden
- Frimodt-Heineke Fonden
- Toyota Foundation
- Tømmerhandler Johannes Fogs Fond
- Dagmar Marshalls Fond
Zitieren
APA
Faurholdt, J., Jacobsen, M. E., Tolsgaard, M. G., & Gustafsson, A. (2026). Error replication as an instructional strategy in simulation‐based procedural learning. Medical Education. https://doi.org/10.1111/medu.70290
Vancouver
Faurholdt J, Jacobsen ME, Tolsgaard MG, Gustafsson A. Error replication as an instructional strategy in simulation‐based procedural learning. Medical Education. 2026. doi:10.1111/medu.70290
BibTeX
@article{Faurholdt2026,
author = {Faurholdt, Jacob and Jacobsen, Mads Emil and Tolsgaard, Martin Grønnebæk and Gustafsson, Amandus},
title = {Error replication as an instructional strategy in simulation‐based procedural learning},
journal = {Medical Education},
year = {2026},
doi = {10.1111/medu.70290}
}
EUVIMED