Use of healthcare claims to validate the Prevention of Arrhythmia Device Infection Trial cardiac implantable electronic device infection risk score

Figures

• 

  Figure 1. Algorithm for initial or replacement CIED procedural group classification. Medical history considered was defined as that occurring within a minimum of 12 months prior to the index CIED procedure. De novo implants or generator replacements of pacemakers, ICDs, or CRT devices were assigned to the Pacemaker, ICD, and CRT categories, respectively, in the determination of the PADIT score. *Data and outliers cleaning process: excluded patients had more than 10 index procedures/dates; excluded records which index procedure is removal only; if records had multiple index procedures dates with same claims ID, they were only considered as one index procedure; for duplicated or overlapped records, they were combined and considered as one index procedure; if patients had multiple index procedures during the same inpatient admission, only the last one was considered, and so on. **If records with missing value for key variables in raw data, for example, place of service variable. ***Place of service in inpatient hospital, outpatient hospital, or ambulatory surgical centre. CIED, cardiac implantable electronic device; CRT, cardiac resynchronization therapy device; ICD, implantable cardiac…

• 

  Figure 2. (A) Annualized rate of major CIED infections by PADIT score. Blue line represents a population similar to the original PADIT risk score analysis (no prior CIED infections) and the orange line represents patients with prior CIED infections. The insert depicts the original PADIT risk score analysis scaled up for clarity. Rates unadjusted for patient, as multiple index procedures per patient were included; only one major CIED infection was allowed per index procedure. (B) Annualized Rate of Major CIED Infections by PADIT score excluding non-de novo first procedures. In the inset, the blue line represents a population similar to the original PADIT risk score analysis (no prior CIED infections) and the orange line represents patients with prior CIED infections. The main figure depicts the original PADIT risk score analysis scaled up for clarity. Non-de novo first procedures were excluded as the number of prior procedures was uncertain owing to lack of this component of the patient history; this analysis was restricted to index procedures for which sufficient patient history was available to determine a PADIT score. The sample size with PADIT score 10–13 was combined due to low…

• 

  Figure 3. (A) Incidence rates of major CIED infection by PADIT score excluding procedures with history of CIED infection. One index procedure was allowed per patient; for patients with more than one procedure the second procedure was used to mitigate the possible underestimation of the PADIT score due to incomplete patient history. (B) Incidence rates of major CIED infection by PADIT score excluding non-de novo first procedures and procedures in patients with prior CIED Infection. Non-de novo first procedures were excluded as the number of prior procedures was uncertain owing to a lack of full patient history; this analysis was restricted to index procedures for which sufficient patient history was available to determine a PADIT score. CIED, cardiac implantable electronic device; PADIT, Prevention of Arrhythmia Device Infection Trial.