BACKGROUND: Delay in the diagnosis of cancer is generally considered unacceptable. However, observational studies often show an inverse association between the length of the diagnostic interval and mortality. Paradoxically, patients diagnosed more rapidly have higher mortality rates than patients with longer wait in the primary and secondary health care sector. AIM: To examine whether the waiting time paradox is manifest in the Danish health care system. MATERIALS & METHODS: The study was based on data on hospital discharge diagnoses for the 2004-2005 period, extracted from population-based healthcare databases in the former County of Aarhus, Denmark. All patients with a first-time diagnosis of breast, skin, lung, prostate, and colorectal cancer were identified and confirmed by each patient’s General Practitioner (GP), who provided a detailed description of the diagnostic pathway. Diagnostic interval was defined as time duration from first presentation of symptom to GP until date of diagnosis. Proportional hazard regression was used to estimate 3-year mortality rate ratios for 15-29 days compared with 0-14, 30-89, and 90+ days of duration, adjusting for gender and age. RESULTS: We identified 1270 cancer patients. Mortality varied across cancer diagnosis. There were no associations between duration and mortality for breast and skin cancer. For lung, prostate, and colorectal cancer a short diagnostic interval (0-14 days) was associated with highest mortality. CONCLUSION: The waiting-time-paradox is manifest in Denmark. We speculate that GPs and hospital doctors are able to distinguish more or less aggressive malignancies and organise the course of referral accordingly.