Background: There have been recent concerns regarding ageism in oncological management. A debate is on-going as to what extent age alone influences the decision to offer a patient adjuvant or palliative chemotherapy, or whether apparent ageism is simply a reflection of correlation between age and other prognostic or predictive factors. This study aims to measure the independent effect of age on the decision to use chemotherapy in 20 common cancers. Methods: Patients with an ICD-10 coded diagnosis of cancer were identified within a regional cancer registry between 2000 and 2011. The data was linked to UK national diagnostic data (Hospital Episode Statistics) to provide information on comorbidity. The odds ratio for receiving chemotherapy dependent on age was estimated using logistic regression to adjust for stage, morphology, grade, socioeconomic status, surgery, radiotherapy, other treatments and 15 different comorbidities. Cox regression analysis on 10 year overall survival was used test for a predictive effect of age on benefit from chemotherapy, adjusted for these factors. Results: 96,165 patients were included in the study. The odds ratio for older patients receiving chemotherapy (age 70 or greater) was 0.03 (95% CI 0.03 - 0.04, p=<0.001) for early breast cancer; 0.10 (95% CI 0.05 – 0.20, p = <0.001) for metastatic breast cancer; 0.43 (95% CI 0.25-0.74, p=0.003) for ovarian cancer; 0.18 (95% CI 0.15 – 0.22, p=<0.001) for Dukes B and C colorectal cancer; 0.21 (95% CI 0.17 – 0.27, p<0.001) for metastatic colorectal cancer and significant for all other cancer types. For overall survival, the interaction between age and chemotherapy was significant (p<0.01) for many tumour types including early breast cancer, colorectal cancer, bladder cancer and pancreatic cancer. Conclusions: Age exerts an independent effect on the decision to treat with chemotherapy in all types of cancer studied. There is evidence that the size of this effect and the age cut-off for a likely decision to treat varies between cancer types. Age may be predictive of benefit from chemotherapy in many cancer types. Further research is needed to clarify to contribution of subjective attributes such as performance status, frailty and patient or clinician attitudes on treatment decisions.