. Author manuscript; available in PMC: 2016 Aug 9.

Published in final edited form as: J Biomed Inform. 2015 Jul 22;58(Suppl):S67–S77. doi: 10.1016/j.jbi.2015.07.001

Table 5.

Smoking status in 2006 and 2014 corpora

Comparison of 2006 and 2014 for smoking status categories
Distribution of smoking status categories			Performance on smoking status categories
	% of 2006 corpus	% of 2014 corpus		Top 2006 F1 scores	Top 2014 F1 scores
Past	9.3	20.1	Past	0.67	.8869
Current	9.2	7.0	Current	0.82	.7143
Non-smoker (never smoked)	16.3	23.3	Non-smoker (never smoked)	0.94	.9237
Smoker (ever smoked)	2.4	.9	Smoker (ever smoked)	0.4	0
Unknown	62.7	47.1	Unknown	0.96	.9688
Unspecified (no consensus reached for gold standard)	-	1.6		-	-