PMID- 33152055 OWN - NLM STAT- MEDLINE DCOM- 20201228 LR - 20201228 IS - 1932-6203 (Electronic) IS - 1932-6203 (Linking) VI - 15 IP - 11 DP - 2020 TI - Measuring adverse events following hip arthroplasty surgery using administrative data without relying on ICD-codes. PG - e0242008 LID - 10.1371/journal.pone.0242008 [doi] LID - e0242008 AB - INTRODUCTION: Measure and monitor adverse events (AEs) following hip arthroplasty is challenging. The aim of this study was to create a model for measuring AEs after hip arthroplasty using administrative data, such as length of stay and readmissions, with equal or better precision than an ICD-code based model. MATERIALS AND METHODS: This study included 1 998 patients operated with an acute or elective hip arthroplasty in a national multi-centre study. We collected AEs within 90 days following surgery with retrospective record review. Additional data came from the Swedish Hip Arthroplasty Register, the Swedish National Patient Register and the Swedish National Board of Health and Welfare. We made a 2:1 split of the data into a training and a holdout set. We used the training set to train different machine learning models to predict if a patient had sustained an AE or not. After training and cross-validation we tested the best performing model on the holdout-set. We compared the results with an established ICD-code based measure for AEs. RESULTS: The best performing model was a logistic regression model with four natural age splines. The variables included in the model were as follows: length of stay at the orthopaedic department, discharge to acute care, age, number of readmissions and ED visits. The sensitivity and specificity for the new model was 23 and 90% for AE within 30 days, compared with 5 and 94% for the ICD-code based model. For AEs within 90 days the sensitivity and specificity were 31% and 89% compared with 16% and 92% for the ICD-code based model. CONCLUSION: We conclude that a prediction model for AEs following hip arthroplasty surgery, relying on administrative data without ICD-codes is more accurate than a model based on ICD-codes. FAU - Magneli, Martin AU - Magneli M AUID- ORCID: 0000-0003-0341-0227 AD - Department of Clinical Sciences, Danderyd Hospital, Division of Orthopaedics, Karolinska Institutet, Stockholm, Sweden. FAU - Unbeck, Maria AU - Unbeck M AD - Department of Neurobiology, Care Sciences and Society, Karolinska Institutet, Stockholm, Sweden. FAU - Rogmark, Cecilia AU - Rogmark C AD - Department of Clinical Sciences, Malmo, Lund University, Lund, Sweden. FAU - Skoldenberg, Olof AU - Skoldenberg O AD - Department of Clinical Sciences, Danderyd Hospital, Division of Orthopaedics, Karolinska Institutet, Stockholm, Sweden. FAU - Gordon, Max AU - Gordon M AUID- ORCID: 0000-0002-8080-5815 AD - Department of Clinical Sciences, Danderyd Hospital, Division of Orthopaedics, Karolinska Institutet, Stockholm, Sweden. LA - eng PT - Journal Article PT - Multicenter Study DEP - 20201105 PL - United States TA - PLoS One JT - PloS one JID - 101285081 SB - IM MH - Adolescent MH - Aged MH - Aged, 80 and over MH - Arthroplasty, Replacement, Hip/*adverse effects MH - Arthroplasty, Replacement, Knee/*adverse effects MH - Female MH - Humans MH - International Classification of Diseases MH - Joints/surgery MH - Length of Stay MH - Logistic Models MH - Male MH - Middle Aged MH - Patient Discharge MH - Retrospective Studies MH - Sensitivity and Specificity MH - Sweden PMC - PMC7644076 COIS- The authors have declared that no competing interests exist. EDAT- 2020/11/06 06:00 MHDA- 2020/12/29 06:00 PMCR- 2020/11/05 CRDT- 2020/11/05 17:14 PHST- 2020/04/07 00:00 [received] PHST- 2020/10/26 00:00 [accepted] PHST- 2020/11/05 17:14 [entrez] PHST- 2020/11/06 06:00 [pubmed] PHST- 2020/12/29 06:00 [medline] PHST- 2020/11/05 00:00 [pmc-release] AID - PONE-D-20-09967 [pii] AID - 10.1371/journal.pone.0242008 [doi] PST - epublish SO - PLoS One. 2020 Nov 5;15(11):e0242008. doi: 10.1371/journal.pone.0242008. eCollection 2020.