OBJECTIVE: The relationship of long-term population-level trends in oral, pharyngeal and oesophageal cancer mortality with major risk factors such as tobacco consumption have not been statistically analysed in Australia. We have demonstrated the long-term implications using historical data. METHODS: Estimated age and sex-specific tobacco consumption back-extrapolated to 1887 were used together with alcohol and fruit and vegetable consumption data to examine their association with trends in oral, pharyngeal and oesophageal cancer mortality. Log-linear Poisson regression models were applied to specify the relationship with oesophageal and pharyngeal mortality data. RESULTS: Oral cancer mortality for males decreased sharply in the first half of the 20th Century in contrast to steadily rising tobacco consumption. Female oral and pharyngeal cancer remained steady at low levels. Post-World War II male and female oesophageal and male pharyngeal cancer mortality rose, then either fell or stabilised, without a clear relationship with risk factors. CONCLUSIONS: Tobacco and alcohol consumption have influenced post-World War II trends in oral, pharyngeal and oesophageal cancer mortality. However, the challenges in using historical population level data prevent precise interpretation of findings. IMPLICATIONS: There is increased exposure to risk factors for these cancers in many low- and middle-income countries. In particular, smoking cessation programs are needed to prevent increases in mortality from these cancers in such countries.