Hair cortisol levels are used to measure long-term stress, while its inactive metabolite cortisone is often not assessed. We measured hair cortisol concentrations (HCC) and hair cortisone concentrations (HCNC) via liquid chromatography quadrupole linear ion trap mass spectrometry (LC-MS) in 62 pregnant women who participated in the LIFE CHILD STUDY in their 2nd and 3rd trimester between 12/2011 and 11/2014. Sociodemographic factors, pregnancy-related factors, and hair characteristics were assessed. Degree of severity of depression, somatization, and stress were evaluated in both trimesters with a self-reported Patient Health Questionnaire (PHQ). Multivariate regression analyses were conducted between HCC and potential influencing factors, as well as with subscales of the PHQ, with HCNC and with the ratio of HCNC to HCC. Spearman correlation coefficients were calculated between steroid concentrations and subscale scores of the PHQ, as well as between the log-fold change in HCC and HCNC and the change in PHQ subscale scores. HCC increased 1.3-fold and HCNC increased 1.5-fold by the 3rd trimester. HCNC was more than three times higher than HCC in both trimesters. We found significant associations of PHQ subscores with HCNC. The PHQ depression score was negatively correlated with HCNC in the 2nd trimester (p < .05). The PHQ stress score change was negatively correlated with the fold change of HCNC (p < .05) and with the change in the ratio of HCNC to HCC (p < .001). Our study suggests an association of cortisol/cortisone metabolism with self-reported stress in the 2nd and 3rd trimester of pregnancy. Since associations with PHQ subscores were only found with cortisone or the ratio of cortisone to cortisol, but not with cortisol alone, both cortisone and cortisol should be used as a marker for stress in pregnant woman.