Purpose: The aim of this study was to evaluate the impact of comprehensive geriatric assessment (CGA)linked to intervention on identified problems on 90-day mortality in older patients with cancer. Patients and methods: Eligible patients were 70 years or older and referred to the Oncology Department at Aarhus University Hospital in order to receive treatment for head and neck cancer (HNC), lung cancer (LC), upper gastrointestinal tract (UGI)cancer or colorectal cancer (CRC). All patients were intendedly invited for CGA. Patients for the study were identified by the oncology department. CGA evaluated six domains: comorbidity, autonomy, mental, cognitive, nutritional status and medication. Intervention was proposed if deficits were detected. Follow-up was performed 90 days after inclusion. Results: From January 1st 2016 through July 31st 2018, 781 patients were identified. Sixty-seven patients were excluded. Median age: 76 (interquartile range: 72–80)years. Primary tumour sites: 5% HNC, 46% LC, 20% UGI and 29% CRC. A total of 407 patients had CGA, 307 had no CGA. Geriatric intervention was proposed in 325 patients (80%)and initiated in 319 patients (78%)in the CGA group. Within 90 days, 142 patients (20%)died. In the non-CGA group, 74 patients died (24%), versus 68 patients (17%)in the CGA group. A potential reduction of death in the CGA group was detected: crude odds ratio (OR): 0.63 (95% confidence interval: 0.43; 0.91), p = 0.014. Adjusted OR: 0.62 (95% confidence interval: 0.39; 1.00), p = 0.05. Conclusion: A CGA linked to oncology evaluation may reduce short-term mortality in older patients with cancer referred for oncological treatment.