OBJECTIVE: The purpose of this study was to determine the risk for separation (ending cohabitation) of the parents of a child with a diagnosis of cancer. METHODS: In a nationwide cohort, we compared the risk for ending cohabitation of the parents of 2450 children (aged 0-20 years) given a diagnosis of cancer with the risk of parents of 44 853 randomly selected, gender- and age-matched cancer-free children. We adjusted for socioeconomic position and demographic factors. Rate ratios and 95% confidence intervals for separation were estimated in a Cox proportional hazards model. RESULTS: The parents of children with cancer did not have a higher risk for separation than the general population (rate ratio: 1.00 [95% confidence interval: 0.91-1.10]). Separate analyses according to type of cancer and survival of the child similarly yielded null results. CONCLUSIONS: Experiencing cancer in a child does not seem to be a risk factor for separation. Our study will allow clinicians to reassure parents and to support them in facing the trauma of cancer in their child.