Smoking And Cancer Mortality: The Authors Reply.

Abstract

Listen

Translate article

Smoking And Cancer Mortality: The Authors Reply We appreciate Frederik Peters’s interest in our article (Apr 2015), and we agree that smoking levels are a strong contributor to mortality rates—both cancer-specific and all-cause. This is one of the reasons why we chose to use the outcome measures we did, and not simply age-adjusted cancer mortality. Peters suggests that the role of smoking and lung cancer rates in driving differences across countries remains strong even after moving from age-adjusted mortality rates to excess and amenablemortality rates, because the latter do not account for background mortality. On the contrary, the very purpose of using excess mortality rates is to account for differences in background mortality rates. Peters is absolutely right that smoking affects not just cancer but allcause mortality rates. For example, smoking is a particularly strong driver of cardiovascular disease. This is why we believe that excess cancer mortality, which estimates the cancer mortality adjusted for backgroundmortality, should be quite effective at controlling for the issue described by Peters. Finally, it should be noted that the list of cancers amenable to health care for the period used in our analysis (1995–2007) did not include any forms of lung cancer. Thus, any differences across countries were entirely unaffected by any differences in changes in lung cancer mortality rates over time. The list of cancers can be seen in table A2 in the online Appendix to our article.