Table 5: An ordered logistic regression model of smoking and adenoma occurrence among individuals ( 𝑛 = 5 , 3 9 5 ) undergoing screening colonoscopy in the New Hampshire Colonscopy Registry from 2004–2006.

Increasing Size C a t e g o r y a of
Adenoma v. No A d e n o m a b
Variable in full m o d e l c Odds Ratio (95% CI)

Smoking
 No pack-years1.00 (referent)
 15 pack-years or fewer1.10 (0.89–1.34)
 More than 15 pack-years1.61 ( 1 . 3 4 - 1 . 9 3 )
Sex
 Male1.00 (referent)
 Female0.54 ( 0 . 4 6 - 0 . 6 3 )
Age (years)1.02 ( 1 . 0 1 - 1 . 0 3 )
Family h i s t o r y d
 Negative1.00 (referent)
 Positive1.69 ( 1 . 3 7 - 2 . 0 8 )

a S i z e categories include < 5 mm, 5–9 mm, 1 cm; odds ratios represent the proportional odds for the logit model. Thus, for one unit change in the predictor variable, the odds for cases in a group less than or equal to a given size category of polyp are the proportional odds times larger.
b T e s t for assumption of proportionality of odds across response categories, using a likelihood ratio test, showed that the assumptions of the ordinal logistic regression were met ( 𝑃 = . 1 9 0 ).
c R e s u l t s reported are odds ratios for each variable in the model, adjusted for all other variables included in the model as shown here.
d F a m i l y history was defined as a reported diagnosis of colon or rectal cancer in a mother, father, sibling, or child of the patient.
𝑃 < . 0 0 1 .