TABLE 3.

Results from regression analyses of melatonin levels by chronotype, night shift workers (NSW) relative to all day shift workers (DSW)

Comparison	% difference in NSW 6- sulfatoxymelatonin levels, relative to DSW levels¶	95% Confidence Interval
Day sleep (NSW) v. night sleep (DSW)
Dichotomous†
Evening	−53.2%**	(−61.1%, −45.3%)
Morning*	−64.6%**	(−70.5%, −58.7%)
Trichotomous§
Evening	−51.3%**	(−60.8%, −41.7%)
Intermediate	−57.2%**	(−64.6%, −49.9%)
Morning*	−71.2%**	(−77.5%, −64.9%)
Night sleep (NSW v. DSW)
Dichotomous†
Evening	−49.0%**	(−57.0%, −41.0%)
Morning*	−30.0%**	(−40.9%, −19.1%)
Trichotomous§
Evening	−49.3%**	(−58.7%, −39.8%)
Intermediate	−38.5%**	(−48.4%, −28.6%)
Morning*	−31.0%**	(−45.2%, −16.8%)
Night work (NSW) v. night sleep (DSW)
Dichotomous†
Evening	−73.2%**	(−77.5%, −69.0%)
Morning*	−54.4%**	(−61.5%, −47.2%)
Trichotomous§
Evening	−76.2%**	(−80.5%, −71.8%)
Intermediate*	−62.2%**	(−68.3%, −56.1%)
Morning*	−51.0%**	(−61.0%, −41.0%)

^*test for difference from Evening Type category: p<0.01, using two-sided t-test

^**p<0.001, two-sided t-test

^†Evening: chronoscore 33 or less; Morning: chronoscore 34 or higher

^§Evening: chronoscore 28 or less; Intermediate: chronoscore 29–39; Morning: chronoscore 40 or higher

^¶Analyzed using the natural log transformation and adjusted for the effects of age, gender, hours of darkness, body mass index, number of alcoholic beverages consumed, and use of psychotheraputics; referent category is all Day Shift Workers (DSW); e.g. in dichotomous analysis, evening type night shift workers during daytime sleep had 53.2% lower levels of 6-sulfatoxmelatonin than all day shift workers during nighttime sleep