The Inaccuracy of ICD-10 Coding in Revision Total Hip Arthroplasty and Its Implication on Revision Data

Lygrisse, Katherine A.; Roof, Mackenzie A.; Keitel, Lauren; Callaghan, John J.; Schwarzkopf, Ran; Bedard, Nicholas A.

doi:10.1016/j.arth.2020.05.013

Cited by 19 publications

(12 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First, we recognize the inherent weaknesses in a large database study including potential for errors in coding and data entry. The transition to using ICD-10 codes in October 2015 was likely associated with increased variations in coding as new norms were being established [ 49 ]. However, our study is one of the first to use ICD-10 codes in a database study evaluating rTHA.…”

Section: Discussionmentioning

confidence: 99%

Instability Is the Most Common Indication for Revision Hip Arthroplasty in the United States: National Trends From 2012 to 2018

et al. 2021

View full text Add to dashboard Cite

Background As primary total hip arthroplasty volume continues to increase, so will the number of revision total hip arthroplasty (rTHA) procedures. These complex cases represent a significant clinical and financial burden to the health-care system. Methods This was a retrospective review using the National Inpatient Sample. International Classification of Diseases, 9th and 10 th revision codes were used to identify patients who underwent rTHA and create cohorts based on rTHA indications from 2012 to 2018. National and regional trends for length of stay (LOS), cost, and discharge location were evaluated. Results A total of 292,250 rTHA procedures were identified. The annual number of rTHA procedures increased by 28.1% from 2012 to 2018 (37,325 to 47,810). The top 3 indications for rTHA were instability (20.4%), aseptic loosening (17.8%), and infection (11.1%). Over the study period, the proportion of patients discharged to skilled nursing facility decreased from 44.2% to 38% ( P < .001). Hospital LOS decreased on average from 4.8 to 4.4 days ( P < .001). Infections had the highest average LOS (7.3 days) followed by periprosthetic fractures (6.5 days). Hospital costs decreased over the study period, from $25,794 to $24,555 ( P < .001). The proportion of rTHA cases performed at urban academic centers increased (58.0% to 75.3%, P < .001) while the proportion performed at urban nonacademic centers decreased (35.5% to 19.4%, P < .001). Conclusion Instability was the most common indication for rTHA between 2012 and 2018. The proportion of rTHA performed in urban academic centers has increased substantially, away from urban nonacademic centers. While cost and LOS have decreased, significant geographic variability exists.

show abstract

Section: Discussionmentioning

confidence: 99%

Instability Is the Most Common Indication for Revision Hip Arthroplasty in the United States: National Trends From 2012 to 2018

et al. 2021

View full text Add to dashboard Cite

show abstract

“…First, we recognize the inherent weaknesses in a large database study including potential for errors in coding and data entry. The transition to using ICD-10 codes in October 2015 was likely associated with increased variations in coding as new norms were being established [ 20 ]. However, our study is one of the first to utilize ICD-10 codes in a database study evaluating rTKA.…”

Section: Discussionmentioning

confidence: 99%

“…On October 1, 2015, the coding standard for the ICD was updated, and the ICD, tenth edition, (ICD-10) was introduced. Since the transition from ICD, ninth editon, to ICD-10, there has been a lack of data describing updated trends in indications for rTKA [ [20] , [21] , [22] , [23] ].…”

Section: Introductionmentioning

confidence: 99%

Epidemiology of Revision Total Knee Arthroplasty in the United States, 2012 to 2019

et al. 2022

View full text Add to dashboard Cite

“…A straightforward way of deriving labels from EHR would be to take advantage of their coded part and use the International Classification of Disease-10th revision-Clinical Modification (referred to as ICD-10, henceforth) codes. This approach has challenges, as demonstrated in multiple studies [2,9,[26][27][28][29][30]. This is especially true if the target information is symptoms, as the corresponding ICD-10 chapter is typically used when a sign or symptom cannot be associated with a definitive diagnosis.…”

Section: Objectivementioning

confidence: 99%

Strategies to Address the Lack of Labeled Data for Supervised Machine Learning Training With Electronic Health Records: Case Study for the Extraction of Symptoms From Clinical Notes (Preprint)

Humbert-Droz¹,

Mukherjee²,

Gevaert³

2021

Preprint

View full text Add to dashboard Cite

BACKGROUND Automated extraction of symptoms from clinical notes is a challenging task owing to the multidimensional nature of symptom description. The availability of labeled training data is extremely limited owing to the nature of the data containing protected health information. Natural language processing and machine learning to process clinical text for such a task have great potential. However, supervised machine learning requires a great amount of labeled data to train a model, which is at the origin of the main bottleneck in model development. OBJECTIVE The aim of this study is to address the lack of labeled data by proposing 2 alternatives to manual labeling for the generation of training labels for supervised machine learning with English clinical text. We aim to demonstrate that using lower-quality labels for training leads to good classification results. METHODS We addressed the lack of labels with 2 strategies. The first approach took advantage of the structured part of electronic health records and used diagnosis codes (International Classification of Disease–10th revision) to derive training labels. The second approach used weak supervision and data programming principles to derive training labels. We propose to apply the developed framework to the extraction of symptom information from outpatient visit progress notes of patients with cardiovascular diseases. RESULTS We used >500,000 notes for training our classification model with International Classification of Disease–10th revision codes as labels and >800,000 notes for training using labels derived from weak supervision. We show that the dependence between prevalence and recall becomes flat provided a sufficiently large training set is used (>500,000 documents). We further demonstrate that using weak labels for training rather than the electronic health record codes derived from the patient encounter leads to an overall improved recall score (10% improvement, on average). Finally, the external validation of our models shows excellent predictive performance and transferability, with an overall increase of 20% in the recall score. CONCLUSIONS This work demonstrates the power of using a weak labeling pipeline to annotate and extract symptom mentions in clinical text, with the prospects to facilitate symptom information integration for a downstream clinical task such as clinical decision support. CLINICALTRIAL

show abstract

The Inaccuracy of ICD-10 Coding in Revision Total Hip Arthroplasty and Its Implication on Revision Data

Cited by 19 publications

References 8 publications

Instability Is the Most Common Indication for Revision Hip Arthroplasty in the United States: National Trends From 2012 to 2018

Instability Is the Most Common Indication for Revision Hip Arthroplasty in the United States: National Trends From 2012 to 2018

Epidemiology of Revision Total Knee Arthroplasty in the United States, 2012 to 2019

Strategies to Address the Lack of Labeled Data for Supervised Machine Learning Training With Electronic Health Records: Case Study for the Extraction of Symptoms From Clinical Notes (Preprint)

Contact Info

Product

Resources

About