Sheldon M Singh1, Lauren Webster, Andrew Calzavara, Harindra C Wijeysundera. 1. *Schulich Heart Centre, Sunnybrook Health Sciences Centre, Faculty of Medicine, University of Toronto †Institute for Clinical Evaluative Sciences (ICES) ‡Institute of Health Policy, Management and Evaluation, University of Toronto §Li Ka Shing Knowledge Institute of St Michael's Hospital, Toronto, ON, Canada.
Abstract
BACKGROUND: Administrative database research can provide insight into the real-world effectiveness of invasive electrophysiology procedures. However, no validated algorithm to identify these procedures within administrative data currently exists. OBJECTIVE: To develop and validate algorithms to identify atrial fibrillation (AF), atrial flutter (AFL), supraventricular tachycardia (SVT) catheter ablation procedures, and diagnostic electrophysiology studies (EPS) within administrative data. METHODS: Algorithms consisting of physician procedural billing codes and their associated most responsible hospital diagnosis codes were used to identify potential AF, AFL, SVT catheter ablation procedures and diagnostic EPS within large administrative databases in Ontario, Canada. The potential procedures were then limited to those performed between October 1, 2011 and March 31, 2013 at a single large regional cardiac center (Sunnybrook Health Sciences Center) in Ontario, Canada. These procedures were compared with a gold-standard cohort of patients known to have undergone invasive electrophysiology procedures during the same time period at the same institution. The sensitivity, specificity, positive and negative predictive values of each algorithm was determined. RESULTS: Algorithms specific to each of AF, AFL, and SVT ablation were associated with a moderate sensitivity (75%-86%), high specificity (95%-98%), positive (95%-98%), and negative (99%) predictive values. The best algorithm to identify diagnostic EPS was less optimal with a sensitivity of 61% and positive predictive value of 88%. CONCLUSIONS: Algorithms using a combination of physician procedural billing codes and accompanying most responsible hospital diagnosis may identify catheter ablation procedures within administrative data with a high degree of accuracy. Diagnostic EPS may be identified with reduced accuracy.
BACKGROUND: Administrative database research can provide insight into the real-world effectiveness of invasive electrophysiology procedures. However, no validated algorithm to identify these procedures within administrative data currently exists. OBJECTIVE: To develop and validate algorithms to identify atrial fibrillation (AF), atrial flutter (AFL), supraventricular tachycardia (SVT) catheter ablation procedures, and diagnostic electrophysiology studies (EPS) within administrative data. METHODS: Algorithms consisting of physician procedural billing codes and their associated most responsible hospital diagnosis codes were used to identify potential AF, AFL, SVT catheter ablation procedures and diagnostic EPS within large administrative databases in Ontario, Canada. The potential procedures were then limited to those performed between October 1, 2011 and March 31, 2013 at a single large regional cardiac center (Sunnybrook Health Sciences Center) in Ontario, Canada. These procedures were compared with a gold-standard cohort of patients known to have undergone invasive electrophysiology procedures during the same time period at the same institution. The sensitivity, specificity, positive and negative predictive values of each algorithm was determined. RESULTS: Algorithms specific to each of AF, AFL, and SVT ablation were associated with a moderate sensitivity (75%-86%), high specificity (95%-98%), positive (95%-98%), and negative (99%) predictive values. The best algorithm to identify diagnostic EPS was less optimal with a sensitivity of 61% and positive predictive value of 88%. CONCLUSIONS: Algorithms using a combination of physician procedural billing codes and accompanying most responsible hospital diagnosis may identify catheter ablation procedures within administrative data with a high degree of accuracy. Diagnostic EPS may be identified with reduced accuracy.
Authors: Paul D James; Mae Hegagi; Lilia Antonova; Jill Tinmouth; Steven J Heitman; Carmine Simone; Elaine Yeung; Elaine Yong Journal: CMAJ Open Date: 2017-06-07
Authors: Idan Roifman; Feng Qiu; Kim A Connelly; Graham A Wright; Michael Farkouh; Laura Jimenez-Juan; Harindra C Wijeysundera Journal: BMJ Open Date: 2018-10-08 Impact factor: 2.692