Skip to main content

Table 1 Smoking status and lung cancer screening eligibility identification in the EHR using an informatics pipeline

From: Assessing data availability and quality within an electronic health record system through external validation against an external clinical data source

Concordance between NHCR reported status and smokers’ registry status

 

Pipeline: Clinical notes only (n = 252)

Pipeline and Semi-structured (n = 400)

Ever smoker vs. Never smoker (n = 403)

Cohen’s Kappa

0.56

0.62

0.59

Concordance

63.9%

67.8%

83.6%

  1. Summary statistics for smoking status detection using the informatics pipeline on clinical notes only, the pipeline merged with semi-structured data, and the merged pipeline with semi-structure data simplified to ever smoker vs. never smoker