We explored predictors of shift work adaptation and how it relates to disease risk biomarker levels. These analyses included 38 male, rotating shift workers, sampled twice at the end of a 3-week night shift and a 3-week day shift rotation. Participants collected all 24-h urine voids, wore activity sensors, and responded to questionnaires during each shift. Using cosinor analysis, we derived the main period of urinary 6-sulfatoxymelatonin (aMT6s) production. Adaptation was defined as the overlap between the main aMT6s production period and sleep period assessed with actigraphy. We used linear models to identify predictors of adaptation to each shift and assessed associations between adaptation profiles and hormone, cytokine, and metabolite biomarker levels. The median duration of overlap (adaptation) was 3.85 h (IQR 2.59-5.03) in the night and 2.98 (IQR 2.17-4.11) in the day shift. In the night shift, a later chronotype (coeff: -1.16, 95% CI -1.87, -0.45) and increased light at night (coeff: -0.97, 95% CI -1.76, -0.18) were associated with poorer adaptation, while longer sleep duration was associated with better adaptation (coeff: 0.46, 95% CI 0.04, 0.88). In the day shift, later sleep onset was associated with worse adaptation (coeff: -0.06, 95% CI -0.12, -0.01), while longer sleep duration was associated with better adaptation (coeff: 0.54, 0.26, 0.81). Results suggest higher androgen and inflammatory marker levels and lower levels of several metabolite markers among less adapted individuals. Chronotype, sleep, and light at night were all associated with night or day shift adaptation. Given the small sample size, results should be viewed as exploratory, but may inform interventions to optimize adaptation of rotating shift workers.