Objectives We aimed to assess the cross-sectional relation between levels of cortisol and specific symptom clusters, symptom severity and duration of symptoms in patients with medically unexplained physical symptoms (MUPS). Methods Baseline data of a cohort of MUPS patients were used. We chose the Cortisol Awakening Response (CAR) as a cortisol parameter, using saliva samples. We used confirmatory factor analysis for the identification of 4 specific symptom clusters: (1) gastro-intestinal symptoms; (2) pain; (3) cardio-pulmonary symptoms; and (4) fatigue. For this factor analysis we used the Physical Symptom Questionnaire (PSQ), which assesses the occurrence and frequency of 51 physical symptoms. Symptom severity was measured with the Patient Health Questionnaire-15 (PHQ-15). Duration of symptoms was based on self-reported duration of top 3 symptoms. We performed multiple linear regression to assess relations between CAR and individual factor scores on symptom clusters, symptom severity and duration of symptoms. Results Data from 296 patients (76% female) were included in the analyses. The majority of patients suffered from symptoms in multiple organ systems. Factor analysis confirmed that the model with 4 symptom clusters fitted our data. For the total study population, we found no significant relation between CAR and participants' factor scores on any of the symptom clusters. We also found no significant relations between CAR and severity or duration of symptoms. Conclusion Our results suggest that within a heterogeneous MUPS population there is no relation between CAR and symptom severity and duration. However, more studies are needed to confirm our findings.